fpadovani/ind-latn-100mb-100mb_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
ind-latn-100mb-100mb_seed3407/checkpoint-9000/trainer_state.json
ModelHub XC 0403ed2c1e 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/ind-latn-100mb-100mb_seed3407
Source: Original Platform
2026-07-07 19:47:52 +08:00

18068 lines
494 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.7002528690916164,
"eval_steps": 3000,
"global_step": 9000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.692063236236573,
"epoch": 0.00038902937171756465,
"grad_norm": 13.75,
"learning_rate": 2e-06,
"loss": 10.8016,
"mean_token_accuracy": 0.0,
"num_tokens": 8755.0,
"step": 5
},
{
"entropy": 10.691958236694337,
"epoch": 0.0007780587434351293,
"grad_norm": 14.0625,
"learning_rate": 4.5e-06,
"loss": 10.7404,
"mean_token_accuracy": 0.000141643057577312,
"num_tokens": 17641.0,
"step": 10
},
{
"entropy": 10.690918159484863,
"epoch": 0.001167088115152694,
"grad_norm": 10.3125,
"learning_rate": 7e-06,
"loss": 10.4828,
"mean_token_accuracy": 0.025441211403813214,
"num_tokens": 26293.0,
"step": 15
},
{
"entropy": 10.67647647857666,
"epoch": 0.0015561174868702586,
"grad_norm": 6.5,
"learning_rate": 9.5e-06,
"loss": 10.2109,
"mean_token_accuracy": 0.040661663934588435,
"num_tokens": 35627.0,
"step": 20
},
{
"entropy": 10.60925121307373,
"epoch": 0.0019451468585878235,
"grad_norm": 4.28125,
"learning_rate": 1.2e-05,
"loss": 9.8983,
"mean_token_accuracy": 0.04408857449889183,
"num_tokens": 43861.0,
"step": 25
},
{
"entropy": 10.547060012817383,
"epoch": 0.002334176230305388,
"grad_norm": 3.625,
"learning_rate": 1.4500000000000002e-05,
"loss": 9.7771,
"mean_token_accuracy": 0.04438098855316639,
"num_tokens": 52816.0,
"step": 30
},
{
"entropy": 10.550695419311523,
"epoch": 0.0027232056020229526,
"grad_norm": 3.046875,
"learning_rate": 1.7000000000000003e-05,
"loss": 9.716,
"mean_token_accuracy": 0.044043789803981784,
"num_tokens": 61370.0,
"step": 35
},
{
"entropy": 10.541547393798828,
"epoch": 0.0031122349737405172,
"grad_norm": 2.84375,
"learning_rate": 1.95e-05,
"loss": 9.6407,
"mean_token_accuracy": 0.04505775421857834,
"num_tokens": 69444.0,
"step": 40
},
{
"entropy": 10.531424808502198,
"epoch": 0.003501264345458082,
"grad_norm": 2.6875,
"learning_rate": 2.2e-05,
"loss": 9.591,
"mean_token_accuracy": 0.04762658141553402,
"num_tokens": 77892.0,
"step": 45
},
{
"entropy": 10.512275695800781,
"epoch": 0.003890293717175647,
"grad_norm": 2.53125,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.5439,
"mean_token_accuracy": 0.061067067086696625,
"num_tokens": 86154.0,
"step": 50
},
{
"entropy": 10.401442527770996,
"epoch": 0.004279323088893211,
"grad_norm": 2.3125,
"learning_rate": 2.7e-05,
"loss": 9.4981,
"mean_token_accuracy": 0.06119193881750107,
"num_tokens": 95589.0,
"step": 55
},
{
"entropy": 10.321024227142335,
"epoch": 0.004668352460610776,
"grad_norm": 2.296875,
"learning_rate": 2.95e-05,
"loss": 9.4607,
"mean_token_accuracy": 0.06384341418743134,
"num_tokens": 105268.0,
"step": 60
},
{
"entropy": 10.332015323638917,
"epoch": 0.0050573818323283405,
"grad_norm": 2.0,
"learning_rate": 3.2e-05,
"loss": 9.2989,
"mean_token_accuracy": 0.06944983527064323,
"num_tokens": 113347.0,
"step": 65
},
{
"entropy": 10.328180885314941,
"epoch": 0.005446411204045905,
"grad_norm": 2.390625,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.2579,
"mean_token_accuracy": 0.06596625149250031,
"num_tokens": 121235.0,
"step": 70
},
{
"entropy": 10.303812599182129,
"epoch": 0.00583544057576347,
"grad_norm": 1.9296875,
"learning_rate": 3.7e-05,
"loss": 9.152,
"mean_token_accuracy": 0.06924093440175057,
"num_tokens": 129529.0,
"step": 75
},
{
"entropy": 10.268609428405762,
"epoch": 0.0062244699474810344,
"grad_norm": 1.9375,
"learning_rate": 3.95e-05,
"loss": 9.0909,
"mean_token_accuracy": 0.06597104258835315,
"num_tokens": 137918.0,
"step": 80
},
{
"entropy": 10.261999797821044,
"epoch": 0.006613499319198599,
"grad_norm": 2.078125,
"learning_rate": 4.2000000000000004e-05,
"loss": 8.9463,
"mean_token_accuracy": 0.06913249380886555,
"num_tokens": 146445.0,
"step": 85
},
{
"entropy": 10.207746601104736,
"epoch": 0.007002528690916164,
"grad_norm": 1.9921875,
"learning_rate": 4.45e-05,
"loss": 8.9607,
"mean_token_accuracy": 0.0626212403178215,
"num_tokens": 155904.0,
"step": 90
},
{
"entropy": 10.237759494781494,
"epoch": 0.007391558062633729,
"grad_norm": 1.8828125,
"learning_rate": 4.7000000000000004e-05,
"loss": 8.8494,
"mean_token_accuracy": 0.06117668002843857,
"num_tokens": 164436.0,
"step": 95
},
{
"entropy": 10.121377658843993,
"epoch": 0.007780587434351294,
"grad_norm": 1.6171875,
"learning_rate": 4.9500000000000004e-05,
"loss": 8.7265,
"mean_token_accuracy": 0.06574496850371361,
"num_tokens": 172985.0,
"step": 100
},
{
"entropy": 10.09178113937378,
"epoch": 0.008169616806068859,
"grad_norm": 1.625,
"learning_rate": 5.2e-05,
"loss": 8.5893,
"mean_token_accuracy": 0.0671191781759262,
"num_tokens": 182084.0,
"step": 105
},
{
"entropy": 10.011064624786377,
"epoch": 0.008558646177786422,
"grad_norm": 1.609375,
"learning_rate": 5.45e-05,
"loss": 8.4644,
"mean_token_accuracy": 0.07035542875528336,
"num_tokens": 190742.0,
"step": 110
},
{
"entropy": 9.856452560424804,
"epoch": 0.008947675549503988,
"grad_norm": 1.609375,
"learning_rate": 5.7e-05,
"loss": 8.3796,
"mean_token_accuracy": 0.06666092537343501,
"num_tokens": 199600.0,
"step": 115
},
{
"entropy": 9.79461441040039,
"epoch": 0.009336704921221552,
"grad_norm": 1.7578125,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.2817,
"mean_token_accuracy": 0.06714933924376965,
"num_tokens": 207883.0,
"step": 120
},
{
"entropy": 9.610589599609375,
"epoch": 0.009725734292939117,
"grad_norm": 1.9453125,
"learning_rate": 6.2e-05,
"loss": 8.2106,
"mean_token_accuracy": 0.06839943863451481,
"num_tokens": 216216.0,
"step": 125
},
{
"entropy": 9.441116333007812,
"epoch": 0.010114763664656681,
"grad_norm": 1.4140625,
"learning_rate": 6.450000000000001e-05,
"loss": 8.0674,
"mean_token_accuracy": 0.06686468720436096,
"num_tokens": 225250.0,
"step": 130
},
{
"entropy": 9.317984580993652,
"epoch": 0.010503793036374246,
"grad_norm": 1.3984375,
"learning_rate": 6.7e-05,
"loss": 7.9861,
"mean_token_accuracy": 0.06643427908420563,
"num_tokens": 233748.0,
"step": 135
},
{
"entropy": 9.106039714813232,
"epoch": 0.01089282240809181,
"grad_norm": 1.578125,
"learning_rate": 6.950000000000001e-05,
"loss": 7.9137,
"mean_token_accuracy": 0.06834710240364075,
"num_tokens": 242629.0,
"step": 140
},
{
"entropy": 9.014894771575928,
"epoch": 0.011281851779809376,
"grad_norm": 1.34375,
"learning_rate": 7.2e-05,
"loss": 7.825,
"mean_token_accuracy": 0.06285446509718895,
"num_tokens": 251458.0,
"step": 145
},
{
"entropy": 8.738780689239501,
"epoch": 0.01167088115152694,
"grad_norm": 1.328125,
"learning_rate": 7.45e-05,
"loss": 7.7458,
"mean_token_accuracy": 0.07414877340197563,
"num_tokens": 260557.0,
"step": 150
},
{
"entropy": 8.657924938201905,
"epoch": 0.012059910523244505,
"grad_norm": 1.1875,
"learning_rate": 7.7e-05,
"loss": 7.7607,
"mean_token_accuracy": 0.06932397484779358,
"num_tokens": 269262.0,
"step": 155
},
{
"entropy": 8.543195724487305,
"epoch": 0.012448939894962069,
"grad_norm": 1.1875,
"learning_rate": 7.950000000000001e-05,
"loss": 7.662,
"mean_token_accuracy": 0.07255172692239284,
"num_tokens": 277804.0,
"step": 160
},
{
"entropy": 8.427197933197021,
"epoch": 0.012837969266679634,
"grad_norm": 1.46875,
"learning_rate": 8.2e-05,
"loss": 7.6187,
"mean_token_accuracy": 0.0732395388185978,
"num_tokens": 287450.0,
"step": 165
},
{
"entropy": 8.372405529022217,
"epoch": 0.013226998638397198,
"grad_norm": 1.125,
"learning_rate": 8.450000000000001e-05,
"loss": 7.6324,
"mean_token_accuracy": 0.07173264883458615,
"num_tokens": 296497.0,
"step": 170
},
{
"entropy": 8.20972490310669,
"epoch": 0.013616028010114764,
"grad_norm": 1.546875,
"learning_rate": 8.7e-05,
"loss": 7.5358,
"mean_token_accuracy": 0.07566781267523766,
"num_tokens": 305414.0,
"step": 175
},
{
"entropy": 8.156252384185791,
"epoch": 0.014005057381832327,
"grad_norm": 1.4140625,
"learning_rate": 8.95e-05,
"loss": 7.5503,
"mean_token_accuracy": 0.07218287587165832,
"num_tokens": 313639.0,
"step": 180
},
{
"entropy": 8.146188354492187,
"epoch": 0.014394086753549893,
"grad_norm": 1.5,
"learning_rate": 9.2e-05,
"loss": 7.5511,
"mean_token_accuracy": 0.07190043628215789,
"num_tokens": 322855.0,
"step": 185
},
{
"entropy": 8.109641361236573,
"epoch": 0.014783116125267459,
"grad_norm": 1.7109375,
"learning_rate": 9.45e-05,
"loss": 7.6463,
"mean_token_accuracy": 0.07696216627955436,
"num_tokens": 332071.0,
"step": 190
},
{
"entropy": 8.094918823242187,
"epoch": 0.015172145496985022,
"grad_norm": 1.3359375,
"learning_rate": 9.7e-05,
"loss": 7.4403,
"mean_token_accuracy": 0.077825241163373,
"num_tokens": 339990.0,
"step": 195
},
{
"entropy": 7.986162710189819,
"epoch": 0.015561174868702588,
"grad_norm": 1.3125,
"learning_rate": 9.95e-05,
"loss": 7.489,
"mean_token_accuracy": 0.0800612710416317,
"num_tokens": 348581.0,
"step": 200
},
{
"entropy": 8.061878490447999,
"epoch": 0.01595020424042015,
"grad_norm": 1.15625,
"learning_rate": 0.000102,
"loss": 7.4901,
"mean_token_accuracy": 0.07649841569364071,
"num_tokens": 357032.0,
"step": 205
},
{
"entropy": 7.995882654190064,
"epoch": 0.016339233612137717,
"grad_norm": 1.28125,
"learning_rate": 0.00010449999999999999,
"loss": 7.4516,
"mean_token_accuracy": 0.0756280355155468,
"num_tokens": 366562.0,
"step": 210
},
{
"entropy": 7.987649202346802,
"epoch": 0.016728262983855283,
"grad_norm": 1.4375,
"learning_rate": 0.000107,
"loss": 7.4428,
"mean_token_accuracy": 0.07756799235939979,
"num_tokens": 374555.0,
"step": 215
},
{
"entropy": 7.977654266357422,
"epoch": 0.017117292355572845,
"grad_norm": 1.40625,
"learning_rate": 0.0001095,
"loss": 7.4725,
"mean_token_accuracy": 0.07444248832762242,
"num_tokens": 383583.0,
"step": 220
},
{
"entropy": 7.902964973449707,
"epoch": 0.01750632172729041,
"grad_norm": 1.2578125,
"learning_rate": 0.000112,
"loss": 7.4967,
"mean_token_accuracy": 0.07443738169968128,
"num_tokens": 392932.0,
"step": 225
},
{
"entropy": 7.966854381561279,
"epoch": 0.017895351099007976,
"grad_norm": 1.3671875,
"learning_rate": 0.0001145,
"loss": 7.3305,
"mean_token_accuracy": 0.07826850302517414,
"num_tokens": 401653.0,
"step": 230
},
{
"entropy": 7.823260259628296,
"epoch": 0.01828438047072554,
"grad_norm": 1.296875,
"learning_rate": 0.00011700000000000001,
"loss": 7.4224,
"mean_token_accuracy": 0.08165781088173389,
"num_tokens": 410432.0,
"step": 235
},
{
"entropy": 7.791707134246826,
"epoch": 0.018673409842443103,
"grad_norm": 1.1015625,
"learning_rate": 0.00011949999999999999,
"loss": 7.3874,
"mean_token_accuracy": 0.07677992731332779,
"num_tokens": 419380.0,
"step": 240
},
{
"entropy": 7.91221809387207,
"epoch": 0.01906243921416067,
"grad_norm": 1.328125,
"learning_rate": 0.000122,
"loss": 7.2983,
"mean_token_accuracy": 0.08029103949666024,
"num_tokens": 427682.0,
"step": 245
},
{
"entropy": 7.782497692108154,
"epoch": 0.019451468585878234,
"grad_norm": 1.140625,
"learning_rate": 0.0001245,
"loss": 7.3944,
"mean_token_accuracy": 0.07596049457788467,
"num_tokens": 437708.0,
"step": 250
},
{
"entropy": 7.839935350418091,
"epoch": 0.0198404979575958,
"grad_norm": 1.2890625,
"learning_rate": 0.000127,
"loss": 7.3625,
"mean_token_accuracy": 0.08329648077487946,
"num_tokens": 446480.0,
"step": 255
},
{
"entropy": 7.786860942840576,
"epoch": 0.020229527329313362,
"grad_norm": 1.3359375,
"learning_rate": 0.0001295,
"loss": 7.3896,
"mean_token_accuracy": 0.0760398305952549,
"num_tokens": 455273.0,
"step": 260
},
{
"entropy": 7.816596269607544,
"epoch": 0.020618556701030927,
"grad_norm": 1.296875,
"learning_rate": 0.000132,
"loss": 7.308,
"mean_token_accuracy": 0.08219457417726517,
"num_tokens": 464306.0,
"step": 265
},
{
"entropy": 7.677759218215942,
"epoch": 0.021007586072748493,
"grad_norm": 1.2265625,
"learning_rate": 0.00013450000000000002,
"loss": 7.3139,
"mean_token_accuracy": 0.0749464474618435,
"num_tokens": 472812.0,
"step": 270
},
{
"entropy": 7.768700742721558,
"epoch": 0.02139661544446606,
"grad_norm": 1.3125,
"learning_rate": 0.00013700000000000002,
"loss": 7.2851,
"mean_token_accuracy": 0.08093390390276908,
"num_tokens": 481326.0,
"step": 275
},
{
"entropy": 7.76422529220581,
"epoch": 0.02178564481618362,
"grad_norm": 1.375,
"learning_rate": 0.0001395,
"loss": 7.2397,
"mean_token_accuracy": 0.08092119097709656,
"num_tokens": 489147.0,
"step": 280
},
{
"entropy": 7.733962726593018,
"epoch": 0.022174674187901186,
"grad_norm": 1.296875,
"learning_rate": 0.00014199999999999998,
"loss": 7.2303,
"mean_token_accuracy": 0.08891455829143524,
"num_tokens": 497455.0,
"step": 285
},
{
"entropy": 7.637256097793579,
"epoch": 0.02256370355961875,
"grad_norm": 1.3671875,
"learning_rate": 0.0001445,
"loss": 7.1567,
"mean_token_accuracy": 0.08741839975118637,
"num_tokens": 506735.0,
"step": 290
},
{
"entropy": 7.682271909713745,
"epoch": 0.022952732931336317,
"grad_norm": 1.296875,
"learning_rate": 0.000147,
"loss": 7.2384,
"mean_token_accuracy": 0.08303680419921874,
"num_tokens": 515976.0,
"step": 295
},
{
"entropy": 7.649052429199219,
"epoch": 0.02334176230305388,
"grad_norm": 1.3515625,
"learning_rate": 0.0001495,
"loss": 7.2294,
"mean_token_accuracy": 0.08583831116557121,
"num_tokens": 524986.0,
"step": 300
},
{
"entropy": 7.687331056594848,
"epoch": 0.023730791674771445,
"grad_norm": 1.390625,
"learning_rate": 0.000152,
"loss": 7.2415,
"mean_token_accuracy": 0.09258508980274201,
"num_tokens": 533146.0,
"step": 305
},
{
"entropy": 7.660231113433838,
"epoch": 0.02411982104648901,
"grad_norm": 1.2734375,
"learning_rate": 0.00015450000000000001,
"loss": 7.1305,
"mean_token_accuracy": 0.09208664670586586,
"num_tokens": 541704.0,
"step": 310
},
{
"entropy": 7.569728803634644,
"epoch": 0.024508850418206576,
"grad_norm": 1.421875,
"learning_rate": 0.000157,
"loss": 7.1132,
"mean_token_accuracy": 0.09219249933958054,
"num_tokens": 550088.0,
"step": 315
},
{
"entropy": 7.553810167312622,
"epoch": 0.024897879789924138,
"grad_norm": 1.375,
"learning_rate": 0.0001595,
"loss": 7.1116,
"mean_token_accuracy": 0.08964145854115486,
"num_tokens": 558763.0,
"step": 320
},
{
"entropy": 7.644041442871094,
"epoch": 0.025286909161641703,
"grad_norm": 1.171875,
"learning_rate": 0.000162,
"loss": 7.2054,
"mean_token_accuracy": 0.08601472824811936,
"num_tokens": 567538.0,
"step": 325
},
{
"entropy": 7.581082773208618,
"epoch": 0.02567593853335927,
"grad_norm": 1.265625,
"learning_rate": 0.00016450000000000001,
"loss": 7.2187,
"mean_token_accuracy": 0.0808300495147705,
"num_tokens": 577626.0,
"step": 330
},
{
"entropy": 7.481959867477417,
"epoch": 0.026064967905076834,
"grad_norm": 1.328125,
"learning_rate": 0.00016700000000000002,
"loss": 7.1528,
"mean_token_accuracy": 0.08528566733002663,
"num_tokens": 587163.0,
"step": 335
},
{
"entropy": 7.5628759384155275,
"epoch": 0.026453997276794396,
"grad_norm": 1.1796875,
"learning_rate": 0.00016950000000000003,
"loss": 7.1352,
"mean_token_accuracy": 0.08886056169867515,
"num_tokens": 595896.0,
"step": 340
},
{
"entropy": 7.6669008255004885,
"epoch": 0.026843026648511962,
"grad_norm": 1.328125,
"learning_rate": 0.00017199999999999998,
"loss": 7.1495,
"mean_token_accuracy": 0.08720074668526649,
"num_tokens": 603815.0,
"step": 345
},
{
"entropy": 7.511917448043823,
"epoch": 0.027232056020229527,
"grad_norm": 1.3125,
"learning_rate": 0.00017449999999999999,
"loss": 7.1631,
"mean_token_accuracy": 0.09270136058330536,
"num_tokens": 612208.0,
"step": 350
},
{
"entropy": 7.517125844955444,
"epoch": 0.027621085391947093,
"grad_norm": 1.21875,
"learning_rate": 0.000177,
"loss": 7.1271,
"mean_token_accuracy": 0.09068955928087234,
"num_tokens": 621840.0,
"step": 355
},
{
"entropy": 7.542174816131592,
"epoch": 0.028010114763664655,
"grad_norm": 1.234375,
"learning_rate": 0.0001795,
"loss": 7.1365,
"mean_token_accuracy": 0.0883019782602787,
"num_tokens": 630770.0,
"step": 360
},
{
"entropy": 7.5013352394104,
"epoch": 0.02839914413538222,
"grad_norm": 1.1796875,
"learning_rate": 0.000182,
"loss": 7.1354,
"mean_token_accuracy": 0.08821968138217925,
"num_tokens": 640313.0,
"step": 365
},
{
"entropy": 7.581635475158691,
"epoch": 0.028788173507099786,
"grad_norm": 1.3046875,
"learning_rate": 0.0001845,
"loss": 7.1571,
"mean_token_accuracy": 0.09343784376978874,
"num_tokens": 648594.0,
"step": 370
},
{
"entropy": 7.465378761291504,
"epoch": 0.02917720287881735,
"grad_norm": 1.5078125,
"learning_rate": 0.000187,
"loss": 7.1645,
"mean_token_accuracy": 0.09112582504749298,
"num_tokens": 656478.0,
"step": 375
},
{
"entropy": 7.526745271682739,
"epoch": 0.029566232250534917,
"grad_norm": 1.1953125,
"learning_rate": 0.0001895,
"loss": 7.1162,
"mean_token_accuracy": 0.09759116843342781,
"num_tokens": 665405.0,
"step": 380
},
{
"entropy": 7.474437618255616,
"epoch": 0.02995526162225248,
"grad_norm": 1.4921875,
"learning_rate": 0.000192,
"loss": 7.1315,
"mean_token_accuracy": 0.09043080434203148,
"num_tokens": 674301.0,
"step": 385
},
{
"entropy": 7.500934886932373,
"epoch": 0.030344290993970045,
"grad_norm": 1.2890625,
"learning_rate": 0.0001945,
"loss": 7.1395,
"mean_token_accuracy": 0.09345336705446243,
"num_tokens": 683159.0,
"step": 390
},
{
"entropy": 7.383391237258911,
"epoch": 0.03073332036568761,
"grad_norm": 1.296875,
"learning_rate": 0.00019700000000000002,
"loss": 6.9546,
"mean_token_accuracy": 0.09936677888035775,
"num_tokens": 690992.0,
"step": 395
},
{
"entropy": 7.346968412399292,
"epoch": 0.031122349737405176,
"grad_norm": 1.3046875,
"learning_rate": 0.00019950000000000002,
"loss": 7.0703,
"mean_token_accuracy": 0.09284654334187507,
"num_tokens": 699934.0,
"step": 400
},
{
"entropy": 7.553390264511108,
"epoch": 0.03151137910912274,
"grad_norm": 1.578125,
"learning_rate": 0.000202,
"loss": 7.1035,
"mean_token_accuracy": 0.09367862120270729,
"num_tokens": 708844.0,
"step": 405
},
{
"entropy": 7.4107770919799805,
"epoch": 0.0319004084808403,
"grad_norm": 1.203125,
"learning_rate": 0.00020449999999999998,
"loss": 7.1629,
"mean_token_accuracy": 0.08833086043596268,
"num_tokens": 718148.0,
"step": 410
},
{
"entropy": 7.559677696228027,
"epoch": 0.03228943785255787,
"grad_norm": 1.359375,
"learning_rate": 0.000207,
"loss": 7.0958,
"mean_token_accuracy": 0.08757410496473313,
"num_tokens": 726859.0,
"step": 415
},
{
"entropy": 7.36133918762207,
"epoch": 0.032678467224275434,
"grad_norm": 1.375,
"learning_rate": 0.0002095,
"loss": 7.1026,
"mean_token_accuracy": 0.09037772119045258,
"num_tokens": 735045.0,
"step": 420
},
{
"entropy": 7.265546607971191,
"epoch": 0.033067496595993,
"grad_norm": 1.09375,
"learning_rate": 0.000212,
"loss": 6.9826,
"mean_token_accuracy": 0.0958486907184124,
"num_tokens": 743538.0,
"step": 425
},
{
"entropy": 7.454471921920776,
"epoch": 0.033456525967710565,
"grad_norm": 1.2421875,
"learning_rate": 0.0002145,
"loss": 7.017,
"mean_token_accuracy": 0.09482243955135346,
"num_tokens": 752518.0,
"step": 430
},
{
"entropy": 7.29300708770752,
"epoch": 0.033845555339428124,
"grad_norm": 1.375,
"learning_rate": 0.00021700000000000002,
"loss": 7.0525,
"mean_token_accuracy": 0.09259525015950203,
"num_tokens": 760354.0,
"step": 435
},
{
"entropy": 7.4215178966522215,
"epoch": 0.03423458471114569,
"grad_norm": 1.3671875,
"learning_rate": 0.0002195,
"loss": 7.1053,
"mean_token_accuracy": 0.09021464213728905,
"num_tokens": 769522.0,
"step": 440
},
{
"entropy": 7.375669765472412,
"epoch": 0.034623614082863255,
"grad_norm": 1.359375,
"learning_rate": 0.000222,
"loss": 7.023,
"mean_token_accuracy": 0.09649827480316162,
"num_tokens": 778548.0,
"step": 445
},
{
"entropy": 7.265977478027343,
"epoch": 0.03501264345458082,
"grad_norm": 1.328125,
"learning_rate": 0.0002245,
"loss": 7.0381,
"mean_token_accuracy": 0.09258587285876274,
"num_tokens": 787666.0,
"step": 450
},
{
"entropy": 7.404888343811035,
"epoch": 0.035401672826298386,
"grad_norm": 1.40625,
"learning_rate": 0.00022700000000000002,
"loss": 6.9548,
"mean_token_accuracy": 0.09554951265454292,
"num_tokens": 796550.0,
"step": 455
},
{
"entropy": 7.379618072509766,
"epoch": 0.03579070219801595,
"grad_norm": 1.5859375,
"learning_rate": 0.00022950000000000002,
"loss": 7.0295,
"mean_token_accuracy": 0.09436309635639191,
"num_tokens": 804549.0,
"step": 460
},
{
"entropy": 7.194296884536743,
"epoch": 0.03617973156973352,
"grad_norm": 1.1640625,
"learning_rate": 0.00023200000000000003,
"loss": 7.0004,
"mean_token_accuracy": 0.09524615257978439,
"num_tokens": 813328.0,
"step": 465
},
{
"entropy": 7.384324455261231,
"epoch": 0.03656876094145108,
"grad_norm": 1.21875,
"learning_rate": 0.00023449999999999998,
"loss": 7.0326,
"mean_token_accuracy": 0.09230693653225899,
"num_tokens": 821733.0,
"step": 470
},
{
"entropy": 7.3139783382415775,
"epoch": 0.03695779031316864,
"grad_norm": 1.234375,
"learning_rate": 0.000237,
"loss": 6.9669,
"mean_token_accuracy": 0.09483846426010131,
"num_tokens": 830243.0,
"step": 475
},
{
"entropy": 7.309220647811889,
"epoch": 0.03734681968488621,
"grad_norm": 1.28125,
"learning_rate": 0.0002395,
"loss": 6.9553,
"mean_token_accuracy": 0.09212375655770302,
"num_tokens": 839059.0,
"step": 480
},
{
"entropy": 7.19914059638977,
"epoch": 0.03773584905660377,
"grad_norm": 1.265625,
"learning_rate": 0.000242,
"loss": 6.8376,
"mean_token_accuracy": 0.10036729499697686,
"num_tokens": 846932.0,
"step": 485
},
{
"entropy": 7.361199140548706,
"epoch": 0.03812487842832134,
"grad_norm": 1.28125,
"learning_rate": 0.0002445,
"loss": 7.0514,
"mean_token_accuracy": 0.08944635763764382,
"num_tokens": 855333.0,
"step": 490
},
{
"entropy": 7.1707884788513185,
"epoch": 0.0385139078000389,
"grad_norm": 1.328125,
"learning_rate": 0.000247,
"loss": 6.809,
"mean_token_accuracy": 0.09626470431685448,
"num_tokens": 863571.0,
"step": 495
},
{
"entropy": 7.152101230621338,
"epoch": 0.03890293717175647,
"grad_norm": 1.21875,
"learning_rate": 0.0002495,
"loss": 6.8996,
"mean_token_accuracy": 0.10517843365669251,
"num_tokens": 872145.0,
"step": 500
},
{
"entropy": 7.187833833694458,
"epoch": 0.039291966543474034,
"grad_norm": 1.1796875,
"learning_rate": 0.000252,
"loss": 6.9394,
"mean_token_accuracy": 0.09892025515437126,
"num_tokens": 880634.0,
"step": 505
},
{
"entropy": 7.235556554794312,
"epoch": 0.0396809959151916,
"grad_norm": 1.359375,
"learning_rate": 0.0002545,
"loss": 6.8712,
"mean_token_accuracy": 0.09984394237399101,
"num_tokens": 889178.0,
"step": 510
},
{
"entropy": 7.149497556686401,
"epoch": 0.04007002528690916,
"grad_norm": 1.34375,
"learning_rate": 0.000257,
"loss": 6.885,
"mean_token_accuracy": 0.09859361201524734,
"num_tokens": 898255.0,
"step": 515
},
{
"entropy": 7.130793571472168,
"epoch": 0.040459054658626724,
"grad_norm": 1.4453125,
"learning_rate": 0.0002595,
"loss": 6.9412,
"mean_token_accuracy": 0.08988485783338547,
"num_tokens": 907369.0,
"step": 520
},
{
"entropy": 7.193523645401001,
"epoch": 0.04084808403034429,
"grad_norm": 1.3046875,
"learning_rate": 0.000262,
"loss": 6.8381,
"mean_token_accuracy": 0.0956579439342022,
"num_tokens": 915946.0,
"step": 525
},
{
"entropy": 7.214367437362671,
"epoch": 0.041237113402061855,
"grad_norm": 1.109375,
"learning_rate": 0.00026450000000000003,
"loss": 6.8968,
"mean_token_accuracy": 0.09464456140995026,
"num_tokens": 925579.0,
"step": 530
},
{
"entropy": 7.194347190856933,
"epoch": 0.04162614277377942,
"grad_norm": 1.1875,
"learning_rate": 0.00026700000000000004,
"loss": 6.8929,
"mean_token_accuracy": 0.09103822857141494,
"num_tokens": 934732.0,
"step": 535
},
{
"entropy": 7.1877021312713625,
"epoch": 0.042015172145496986,
"grad_norm": 1.390625,
"learning_rate": 0.00026950000000000005,
"loss": 6.9548,
"mean_token_accuracy": 0.09189340770244599,
"num_tokens": 943371.0,
"step": 540
},
{
"entropy": 7.129611682891846,
"epoch": 0.04240420151721455,
"grad_norm": 1.28125,
"learning_rate": 0.00027200000000000005,
"loss": 6.7935,
"mean_token_accuracy": 0.09684587121009827,
"num_tokens": 951977.0,
"step": 545
},
{
"entropy": 7.075324392318725,
"epoch": 0.04279323088893212,
"grad_norm": 1.1875,
"learning_rate": 0.0002745,
"loss": 6.7956,
"mean_token_accuracy": 0.09852224737405776,
"num_tokens": 960474.0,
"step": 550
},
{
"entropy": 7.242029809951783,
"epoch": 0.04318226026064968,
"grad_norm": 1.1796875,
"learning_rate": 0.000277,
"loss": 6.921,
"mean_token_accuracy": 0.09615759328007698,
"num_tokens": 969823.0,
"step": 555
},
{
"entropy": 7.148642349243164,
"epoch": 0.04357128963236724,
"grad_norm": 1.421875,
"learning_rate": 0.0002795,
"loss": 6.8541,
"mean_token_accuracy": 0.10372538045048714,
"num_tokens": 977417.0,
"step": 560
},
{
"entropy": 7.050511121749878,
"epoch": 0.04396031900408481,
"grad_norm": 1.3671875,
"learning_rate": 0.00028199999999999997,
"loss": 6.8525,
"mean_token_accuracy": 0.10765723884105682,
"num_tokens": 985519.0,
"step": 565
},
{
"entropy": 7.106280851364136,
"epoch": 0.04434934837580237,
"grad_norm": 1.546875,
"learning_rate": 0.0002845,
"loss": 6.8486,
"mean_token_accuracy": 0.10191520527005196,
"num_tokens": 994136.0,
"step": 570
},
{
"entropy": 7.03365535736084,
"epoch": 0.04473837774751994,
"grad_norm": 1.3359375,
"learning_rate": 0.000287,
"loss": 6.8207,
"mean_token_accuracy": 0.09769327566027641,
"num_tokens": 1002896.0,
"step": 575
},
{
"entropy": 7.198338603973388,
"epoch": 0.0451274071192375,
"grad_norm": 1.34375,
"learning_rate": 0.0002895,
"loss": 6.9061,
"mean_token_accuracy": 0.10421423837542534,
"num_tokens": 1011032.0,
"step": 580
},
{
"entropy": 7.051746559143067,
"epoch": 0.04551643649095507,
"grad_norm": 1.140625,
"learning_rate": 0.000292,
"loss": 6.8708,
"mean_token_accuracy": 0.09394787698984146,
"num_tokens": 1019901.0,
"step": 585
},
{
"entropy": 7.067822074890136,
"epoch": 0.045905465862672634,
"grad_norm": 1.21875,
"learning_rate": 0.0002945,
"loss": 6.8482,
"mean_token_accuracy": 0.09919805154204368,
"num_tokens": 1029049.0,
"step": 590
},
{
"entropy": 6.9990034103393555,
"epoch": 0.0462944952343902,
"grad_norm": 1.25,
"learning_rate": 0.000297,
"loss": 6.8086,
"mean_token_accuracy": 0.10822348445653915,
"num_tokens": 1037580.0,
"step": 595
},
{
"entropy": 7.196728610992432,
"epoch": 0.04668352460610776,
"grad_norm": 1.2421875,
"learning_rate": 0.0002995,
"loss": 6.8679,
"mean_token_accuracy": 0.10288131684064865,
"num_tokens": 1045804.0,
"step": 600
},
{
"entropy": 7.0782746315002445,
"epoch": 0.047072553977825324,
"grad_norm": 1.2421875,
"learning_rate": 0.000302,
"loss": 6.8362,
"mean_token_accuracy": 0.09710147231817245,
"num_tokens": 1054719.0,
"step": 605
},
{
"entropy": 7.09404673576355,
"epoch": 0.04746158334954289,
"grad_norm": 1.328125,
"learning_rate": 0.0003045,
"loss": 6.8149,
"mean_token_accuracy": 0.1012077234685421,
"num_tokens": 1063373.0,
"step": 610
},
{
"entropy": 6.984814739227295,
"epoch": 0.047850612721260455,
"grad_norm": 1.1953125,
"learning_rate": 0.000307,
"loss": 6.8058,
"mean_token_accuracy": 0.10036636292934417,
"num_tokens": 1073021.0,
"step": 615
},
{
"entropy": 7.0786560535430905,
"epoch": 0.04823964209297802,
"grad_norm": 1.3125,
"learning_rate": 0.0003095,
"loss": 6.8696,
"mean_token_accuracy": 0.09600509628653527,
"num_tokens": 1081547.0,
"step": 620
},
{
"entropy": 7.0220740795135494,
"epoch": 0.048628671464695586,
"grad_norm": 1.15625,
"learning_rate": 0.000312,
"loss": 6.8233,
"mean_token_accuracy": 0.09918136522173882,
"num_tokens": 1090695.0,
"step": 625
},
{
"entropy": 6.9606041431427,
"epoch": 0.04901770083641315,
"grad_norm": 1.2109375,
"learning_rate": 0.0003145,
"loss": 6.7198,
"mean_token_accuracy": 0.10607226341962814,
"num_tokens": 1099563.0,
"step": 630
},
{
"entropy": 7.042634630203247,
"epoch": 0.04940673020813072,
"grad_norm": 1.15625,
"learning_rate": 0.000317,
"loss": 6.8025,
"mean_token_accuracy": 0.09708617925643921,
"num_tokens": 1109021.0,
"step": 635
},
{
"entropy": 7.036008024215699,
"epoch": 0.049795759579848276,
"grad_norm": 1.2734375,
"learning_rate": 0.0003195,
"loss": 6.7915,
"mean_token_accuracy": 0.10409486815333366,
"num_tokens": 1118281.0,
"step": 640
},
{
"entropy": 7.0259270668029785,
"epoch": 0.05018478895156584,
"grad_norm": 1.2421875,
"learning_rate": 0.000322,
"loss": 6.8364,
"mean_token_accuracy": 0.09575799703598023,
"num_tokens": 1127513.0,
"step": 645
},
{
"entropy": 6.843554496765137,
"epoch": 0.05057381832328341,
"grad_norm": 1.296875,
"learning_rate": 0.00032450000000000003,
"loss": 6.7179,
"mean_token_accuracy": 0.10220488980412483,
"num_tokens": 1136140.0,
"step": 650
},
{
"entropy": 7.010839796066284,
"epoch": 0.05096284769500097,
"grad_norm": 1.1875,
"learning_rate": 0.00032700000000000003,
"loss": 6.7449,
"mean_token_accuracy": 0.10034143850207329,
"num_tokens": 1144297.0,
"step": 655
},
{
"entropy": 7.005735111236572,
"epoch": 0.05135187706671854,
"grad_norm": 1.3046875,
"learning_rate": 0.00032950000000000004,
"loss": 6.8315,
"mean_token_accuracy": 0.09598958343267441,
"num_tokens": 1153639.0,
"step": 660
},
{
"entropy": 7.042137622833252,
"epoch": 0.0517409064384361,
"grad_norm": 1.1796875,
"learning_rate": 0.00033200000000000005,
"loss": 6.8044,
"mean_token_accuracy": 0.09692039787769317,
"num_tokens": 1162825.0,
"step": 665
},
{
"entropy": 6.893413639068603,
"epoch": 0.05212993581015367,
"grad_norm": 1.2734375,
"learning_rate": 0.00033450000000000005,
"loss": 6.8242,
"mean_token_accuracy": 0.10342787951231003,
"num_tokens": 1171887.0,
"step": 670
},
{
"entropy": 6.990288639068604,
"epoch": 0.052518965181871234,
"grad_norm": 1.28125,
"learning_rate": 0.000337,
"loss": 6.723,
"mean_token_accuracy": 0.10218419954180717,
"num_tokens": 1179963.0,
"step": 675
},
{
"entropy": 6.915320777893067,
"epoch": 0.05290799455358879,
"grad_norm": 1.2578125,
"learning_rate": 0.0003395,
"loss": 6.6987,
"mean_token_accuracy": 0.10984139442443848,
"num_tokens": 1188408.0,
"step": 680
},
{
"entropy": 6.944253206253052,
"epoch": 0.05329702392530636,
"grad_norm": 1.1875,
"learning_rate": 0.000342,
"loss": 6.7404,
"mean_token_accuracy": 0.09963250234723091,
"num_tokens": 1197751.0,
"step": 685
},
{
"entropy": 6.910943222045899,
"epoch": 0.053686053297023924,
"grad_norm": 1.2421875,
"learning_rate": 0.00034449999999999997,
"loss": 6.8019,
"mean_token_accuracy": 0.10173860564827919,
"num_tokens": 1207101.0,
"step": 690
},
{
"entropy": 6.96247820854187,
"epoch": 0.05407508266874149,
"grad_norm": 1.078125,
"learning_rate": 0.000347,
"loss": 6.7768,
"mean_token_accuracy": 0.10386782586574554,
"num_tokens": 1216395.0,
"step": 695
},
{
"entropy": 6.914447498321533,
"epoch": 0.054464112040459055,
"grad_norm": 1.265625,
"learning_rate": 0.0003495,
"loss": 6.7579,
"mean_token_accuracy": 0.10488244742155076,
"num_tokens": 1225496.0,
"step": 700
},
{
"entropy": 6.9863011837005615,
"epoch": 0.05485314141217662,
"grad_norm": 1.21875,
"learning_rate": 0.000352,
"loss": 6.8255,
"mean_token_accuracy": 0.10280769392848015,
"num_tokens": 1233934.0,
"step": 705
},
{
"entropy": 6.947722387313843,
"epoch": 0.055242170783894186,
"grad_norm": 1.4140625,
"learning_rate": 0.0003545,
"loss": 6.6909,
"mean_token_accuracy": 0.10623404085636139,
"num_tokens": 1242165.0,
"step": 710
},
{
"entropy": 6.864742231369019,
"epoch": 0.05563120015561175,
"grad_norm": 1.3203125,
"learning_rate": 0.000357,
"loss": 6.7451,
"mean_token_accuracy": 0.10805843472480774,
"num_tokens": 1250021.0,
"step": 715
},
{
"entropy": 6.967834186553955,
"epoch": 0.05602022952732931,
"grad_norm": 1.15625,
"learning_rate": 0.0003595,
"loss": 6.7789,
"mean_token_accuracy": 0.1014004610478878,
"num_tokens": 1258397.0,
"step": 720
},
{
"entropy": 6.895936393737793,
"epoch": 0.056409258899046875,
"grad_norm": 1.25,
"learning_rate": 0.000362,
"loss": 6.7594,
"mean_token_accuracy": 0.10026679039001465,
"num_tokens": 1266764.0,
"step": 725
},
{
"entropy": 6.976710271835327,
"epoch": 0.05679828827076444,
"grad_norm": 1.2265625,
"learning_rate": 0.0003645,
"loss": 6.7604,
"mean_token_accuracy": 0.10308134779334069,
"num_tokens": 1275643.0,
"step": 730
},
{
"entropy": 6.854802751541138,
"epoch": 0.057187317642482006,
"grad_norm": 1.21875,
"learning_rate": 0.000367,
"loss": 6.7414,
"mean_token_accuracy": 0.10123410001397133,
"num_tokens": 1284379.0,
"step": 735
},
{
"entropy": 6.858791494369507,
"epoch": 0.05757634701419957,
"grad_norm": 1.2578125,
"learning_rate": 0.0003695,
"loss": 6.5942,
"mean_token_accuracy": 0.10925214067101478,
"num_tokens": 1293115.0,
"step": 740
},
{
"entropy": 6.892751264572143,
"epoch": 0.05796537638591714,
"grad_norm": 1.296875,
"learning_rate": 0.000372,
"loss": 6.796,
"mean_token_accuracy": 0.09836092367768287,
"num_tokens": 1301583.0,
"step": 745
},
{
"entropy": 6.954816818237305,
"epoch": 0.0583544057576347,
"grad_norm": 1.1171875,
"learning_rate": 0.0003745,
"loss": 6.7001,
"mean_token_accuracy": 0.10808140113949775,
"num_tokens": 1310721.0,
"step": 750
},
{
"entropy": 6.923498821258545,
"epoch": 0.05874343512935227,
"grad_norm": 1.4140625,
"learning_rate": 0.000377,
"loss": 6.8595,
"mean_token_accuracy": 0.09595380201935769,
"num_tokens": 1319056.0,
"step": 755
},
{
"entropy": 6.839234256744385,
"epoch": 0.059132464501069834,
"grad_norm": 1.1875,
"learning_rate": 0.0003795,
"loss": 6.7416,
"mean_token_accuracy": 0.10261751636862755,
"num_tokens": 1327259.0,
"step": 760
},
{
"entropy": 6.819512987136841,
"epoch": 0.05952149387278739,
"grad_norm": 1.28125,
"learning_rate": 0.000382,
"loss": 6.6414,
"mean_token_accuracy": 0.10097026452422142,
"num_tokens": 1335906.0,
"step": 765
},
{
"entropy": 6.879472923278809,
"epoch": 0.05991052324450496,
"grad_norm": 1.140625,
"learning_rate": 0.0003845,
"loss": 6.6938,
"mean_token_accuracy": 0.09806213229894638,
"num_tokens": 1344743.0,
"step": 770
},
{
"entropy": 6.840222930908203,
"epoch": 0.060299552616222524,
"grad_norm": 1.234375,
"learning_rate": 0.00038700000000000003,
"loss": 6.6834,
"mean_token_accuracy": 0.10951419249176979,
"num_tokens": 1353211.0,
"step": 775
},
{
"entropy": 6.946290063858032,
"epoch": 0.06068858198794009,
"grad_norm": 1.1796875,
"learning_rate": 0.00038950000000000003,
"loss": 6.812,
"mean_token_accuracy": 0.10016649737954139,
"num_tokens": 1361849.0,
"step": 780
},
{
"entropy": 6.8946840286254885,
"epoch": 0.061077611359657655,
"grad_norm": 1.46875,
"learning_rate": 0.00039200000000000004,
"loss": 6.7945,
"mean_token_accuracy": 0.10707032829523086,
"num_tokens": 1369804.0,
"step": 785
},
{
"entropy": 6.7396095275878904,
"epoch": 0.06146664073137522,
"grad_norm": 1.1171875,
"learning_rate": 0.00039450000000000005,
"loss": 6.5766,
"mean_token_accuracy": 0.10397438257932663,
"num_tokens": 1378478.0,
"step": 790
},
{
"entropy": 6.9116943359375,
"epoch": 0.061855670103092786,
"grad_norm": 1.2890625,
"learning_rate": 0.00039700000000000005,
"loss": 6.6789,
"mean_token_accuracy": 0.10764730274677277,
"num_tokens": 1387225.0,
"step": 795
},
{
"entropy": 6.7650645732879635,
"epoch": 0.06224469947481035,
"grad_norm": 1.265625,
"learning_rate": 0.0003995,
"loss": 6.6492,
"mean_token_accuracy": 0.10592131912708283,
"num_tokens": 1395338.0,
"step": 800
},
{
"entropy": 6.821805953979492,
"epoch": 0.06263372884652792,
"grad_norm": 1.1640625,
"learning_rate": 0.000402,
"loss": 6.63,
"mean_token_accuracy": 0.10603216290473938,
"num_tokens": 1403794.0,
"step": 805
},
{
"entropy": 6.708382701873779,
"epoch": 0.06302275821824548,
"grad_norm": 1.109375,
"learning_rate": 0.0004045,
"loss": 6.7133,
"mean_token_accuracy": 0.10304158627986908,
"num_tokens": 1412522.0,
"step": 810
},
{
"entropy": 6.788723707199097,
"epoch": 0.06341178758996305,
"grad_norm": 1.15625,
"learning_rate": 0.00040699999999999997,
"loss": 6.6937,
"mean_token_accuracy": 0.10893446058034897,
"num_tokens": 1421347.0,
"step": 815
},
{
"entropy": 6.8365006923675535,
"epoch": 0.0638008169616806,
"grad_norm": 1.234375,
"learning_rate": 0.0004095,
"loss": 6.6695,
"mean_token_accuracy": 0.11350144445896149,
"num_tokens": 1429788.0,
"step": 820
},
{
"entropy": 6.760780572891235,
"epoch": 0.06418984633339817,
"grad_norm": 1.3046875,
"learning_rate": 0.000412,
"loss": 6.6727,
"mean_token_accuracy": 0.10774989053606987,
"num_tokens": 1437442.0,
"step": 825
},
{
"entropy": 6.757859325408935,
"epoch": 0.06457887570511574,
"grad_norm": 1.09375,
"learning_rate": 0.0004145,
"loss": 6.5954,
"mean_token_accuracy": 0.10944118574261666,
"num_tokens": 1445604.0,
"step": 830
},
{
"entropy": 6.786571073532104,
"epoch": 0.0649679050768333,
"grad_norm": 1.1875,
"learning_rate": 0.000417,
"loss": 6.6634,
"mean_token_accuracy": 0.1089187815785408,
"num_tokens": 1454342.0,
"step": 835
},
{
"entropy": 6.77791576385498,
"epoch": 0.06535693444855087,
"grad_norm": 1.1875,
"learning_rate": 0.0004195,
"loss": 6.661,
"mean_token_accuracy": 0.11053230538964272,
"num_tokens": 1463189.0,
"step": 840
},
{
"entropy": 6.787366485595703,
"epoch": 0.06574596382026843,
"grad_norm": 1.203125,
"learning_rate": 0.000422,
"loss": 6.7166,
"mean_token_accuracy": 0.10734177976846696,
"num_tokens": 1471608.0,
"step": 845
},
{
"entropy": 6.799466276168824,
"epoch": 0.066134993191986,
"grad_norm": 1.3125,
"learning_rate": 0.0004245,
"loss": 6.7296,
"mean_token_accuracy": 0.10613490417599677,
"num_tokens": 1480452.0,
"step": 850
},
{
"entropy": 6.822136163711548,
"epoch": 0.06652402256370356,
"grad_norm": 1.2421875,
"learning_rate": 0.000427,
"loss": 6.6437,
"mean_token_accuracy": 0.10718747153878212,
"num_tokens": 1489118.0,
"step": 855
},
{
"entropy": 6.686193990707397,
"epoch": 0.06691305193542113,
"grad_norm": 1.2421875,
"learning_rate": 0.0004295,
"loss": 6.6342,
"mean_token_accuracy": 0.10919305682182312,
"num_tokens": 1498961.0,
"step": 860
},
{
"entropy": 6.819766521453857,
"epoch": 0.06730208130713869,
"grad_norm": 1.2421875,
"learning_rate": 0.000432,
"loss": 6.6401,
"mean_token_accuracy": 0.11007228568196296,
"num_tokens": 1506772.0,
"step": 865
},
{
"entropy": 6.681567525863647,
"epoch": 0.06769111067885625,
"grad_norm": 1.2734375,
"learning_rate": 0.0004345,
"loss": 6.5275,
"mean_token_accuracy": 0.11715472564101219,
"num_tokens": 1515095.0,
"step": 870
},
{
"entropy": 6.6687061309814455,
"epoch": 0.06808014005057382,
"grad_norm": 1.1484375,
"learning_rate": 0.000437,
"loss": 6.5653,
"mean_token_accuracy": 0.11424972265958785,
"num_tokens": 1523477.0,
"step": 875
},
{
"entropy": 6.689616775512695,
"epoch": 0.06846916942229138,
"grad_norm": 1.1015625,
"learning_rate": 0.0004395,
"loss": 6.6037,
"mean_token_accuracy": 0.10498946905136108,
"num_tokens": 1531786.0,
"step": 880
},
{
"entropy": 6.736538553237915,
"epoch": 0.06885819879400895,
"grad_norm": 1.171875,
"learning_rate": 0.000442,
"loss": 6.7016,
"mean_token_accuracy": 0.10352547094225883,
"num_tokens": 1540737.0,
"step": 885
},
{
"entropy": 6.637676811218261,
"epoch": 0.06924722816572651,
"grad_norm": 1.1953125,
"learning_rate": 0.0004445,
"loss": 6.5418,
"mean_token_accuracy": 0.1120985560119152,
"num_tokens": 1548923.0,
"step": 890
},
{
"entropy": 6.724155473709106,
"epoch": 0.06963625753744408,
"grad_norm": 1.375,
"learning_rate": 0.000447,
"loss": 6.5185,
"mean_token_accuracy": 0.11639119386672973,
"num_tokens": 1557043.0,
"step": 895
},
{
"entropy": 6.5780919075012205,
"epoch": 0.07002528690916164,
"grad_norm": 1.2890625,
"learning_rate": 0.00044950000000000003,
"loss": 6.5535,
"mean_token_accuracy": 0.11611853837966919,
"num_tokens": 1564994.0,
"step": 900
},
{
"entropy": 6.693197536468506,
"epoch": 0.0704143162808792,
"grad_norm": 1.171875,
"learning_rate": 0.00045200000000000004,
"loss": 6.5559,
"mean_token_accuracy": 0.11024467647075653,
"num_tokens": 1573859.0,
"step": 905
},
{
"entropy": 6.661924743652344,
"epoch": 0.07080334565259677,
"grad_norm": 1.1328125,
"learning_rate": 0.00045450000000000004,
"loss": 6.5138,
"mean_token_accuracy": 0.10926207825541497,
"num_tokens": 1581947.0,
"step": 910
},
{
"entropy": 6.66906771659851,
"epoch": 0.07119237502431433,
"grad_norm": 1.15625,
"learning_rate": 0.00045700000000000005,
"loss": 6.4995,
"mean_token_accuracy": 0.11228668987751007,
"num_tokens": 1589971.0,
"step": 915
},
{
"entropy": 6.555896234512329,
"epoch": 0.0715814043960319,
"grad_norm": 1.21875,
"learning_rate": 0.00045950000000000006,
"loss": 6.4518,
"mean_token_accuracy": 0.11207407787442207,
"num_tokens": 1598612.0,
"step": 920
},
{
"entropy": 6.718100690841675,
"epoch": 0.07197043376774946,
"grad_norm": 1.1328125,
"learning_rate": 0.000462,
"loss": 6.6111,
"mean_token_accuracy": 0.10551391988992691,
"num_tokens": 1607378.0,
"step": 925
},
{
"entropy": 6.653341817855835,
"epoch": 0.07235946313946703,
"grad_norm": 1.078125,
"learning_rate": 0.0004645,
"loss": 6.6081,
"mean_token_accuracy": 0.11038231924176216,
"num_tokens": 1615987.0,
"step": 930
},
{
"entropy": 6.486403131484986,
"epoch": 0.07274849251118459,
"grad_norm": 1.125,
"learning_rate": 0.000467,
"loss": 6.4005,
"mean_token_accuracy": 0.11754591017961502,
"num_tokens": 1625366.0,
"step": 935
},
{
"entropy": 6.682313919067383,
"epoch": 0.07313752188290217,
"grad_norm": 1.1015625,
"learning_rate": 0.0004695,
"loss": 6.5758,
"mean_token_accuracy": 0.11200849413871765,
"num_tokens": 1633020.0,
"step": 940
},
{
"entropy": 6.592645740509033,
"epoch": 0.07352655125461972,
"grad_norm": 1.1484375,
"learning_rate": 0.000472,
"loss": 6.4715,
"mean_token_accuracy": 0.11014393791556358,
"num_tokens": 1641518.0,
"step": 945
},
{
"entropy": 6.5744109630584715,
"epoch": 0.07391558062633728,
"grad_norm": 1.125,
"learning_rate": 0.0004745,
"loss": 6.5601,
"mean_token_accuracy": 0.1099576011300087,
"num_tokens": 1650472.0,
"step": 950
},
{
"entropy": 6.645666933059692,
"epoch": 0.07430460999805485,
"grad_norm": 1.1796875,
"learning_rate": 0.000477,
"loss": 6.5753,
"mean_token_accuracy": 0.11195494011044502,
"num_tokens": 1659037.0,
"step": 955
},
{
"entropy": 6.697547388076782,
"epoch": 0.07469363936977241,
"grad_norm": 1.1875,
"learning_rate": 0.0004795,
"loss": 6.6649,
"mean_token_accuracy": 0.11236765310168266,
"num_tokens": 1667897.0,
"step": 960
},
{
"entropy": 6.552785778045655,
"epoch": 0.07508266874148999,
"grad_norm": 1.1484375,
"learning_rate": 0.000482,
"loss": 6.5075,
"mean_token_accuracy": 0.10858357399702072,
"num_tokens": 1676535.0,
"step": 965
},
{
"entropy": 6.665571069717407,
"epoch": 0.07547169811320754,
"grad_norm": 1.1796875,
"learning_rate": 0.0004845,
"loss": 6.5258,
"mean_token_accuracy": 0.11193247660994529,
"num_tokens": 1684415.0,
"step": 970
},
{
"entropy": 6.619773483276367,
"epoch": 0.07586072748492512,
"grad_norm": 1.25,
"learning_rate": 0.000487,
"loss": 6.4873,
"mean_token_accuracy": 0.1161974124610424,
"num_tokens": 1692443.0,
"step": 975
},
{
"entropy": 6.469988203048706,
"epoch": 0.07624975685664268,
"grad_norm": 1.1015625,
"learning_rate": 0.0004895,
"loss": 6.4908,
"mean_token_accuracy": 0.11075597926974297,
"num_tokens": 1700814.0,
"step": 980
},
{
"entropy": 6.469910621643066,
"epoch": 0.07663878622836025,
"grad_norm": 1.1328125,
"learning_rate": 0.000492,
"loss": 6.3349,
"mean_token_accuracy": 0.1179271623492241,
"num_tokens": 1709591.0,
"step": 985
},
{
"entropy": 6.605605697631836,
"epoch": 0.0770278156000778,
"grad_norm": 1.046875,
"learning_rate": 0.0004945,
"loss": 6.5879,
"mean_token_accuracy": 0.10372007116675377,
"num_tokens": 1718951.0,
"step": 990
},
{
"entropy": 6.66983380317688,
"epoch": 0.07741684497179536,
"grad_norm": 1.1953125,
"learning_rate": 0.000497,
"loss": 6.5431,
"mean_token_accuracy": 0.11147809475660324,
"num_tokens": 1727481.0,
"step": 995
},
{
"entropy": 6.5515549659729,
"epoch": 0.07780587434351294,
"grad_norm": 1.09375,
"learning_rate": 0.0004995,
"loss": 6.5418,
"mean_token_accuracy": 0.11638221740722657,
"num_tokens": 1736985.0,
"step": 1000
},
{
"entropy": 6.616484832763672,
"epoch": 0.0781949037152305,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999999989075146,
"loss": 6.5945,
"mean_token_accuracy": 0.10921485722064972,
"num_tokens": 1745470.0,
"step": 1005
},
{
"entropy": 6.620965623855591,
"epoch": 0.07858393308694807,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999999944692927,
"loss": 6.5176,
"mean_token_accuracy": 0.11291603669524193,
"num_tokens": 1754185.0,
"step": 1010
},
{
"entropy": 6.570215940475464,
"epoch": 0.07897296245866563,
"grad_norm": 1.1796875,
"learning_rate": 0.000499999986617054,
"loss": 6.5153,
"mean_token_accuracy": 0.11022635996341705,
"num_tokens": 1762100.0,
"step": 1015
},
{
"entropy": 6.5821075439453125,
"epoch": 0.0793619918303832,
"grad_norm": 1.171875,
"learning_rate": 0.0004999999753507986,
"loss": 6.5529,
"mean_token_accuracy": 0.11068348959088326,
"num_tokens": 1771070.0,
"step": 1020
},
{
"entropy": 6.546429920196533,
"epoch": 0.07975102120210076,
"grad_norm": 1.109375,
"learning_rate": 0.0004999999606705267,
"loss": 6.5267,
"mean_token_accuracy": 0.11509618312120437,
"num_tokens": 1780085.0,
"step": 1025
},
{
"entropy": 6.625613689422607,
"epoch": 0.08014005057381832,
"grad_norm": 1.03125,
"learning_rate": 0.0004999999425762385,
"loss": 6.5244,
"mean_token_accuracy": 0.11561840996146203,
"num_tokens": 1789316.0,
"step": 1030
},
{
"entropy": 6.537080764770508,
"epoch": 0.08052907994553589,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999999210679344,
"loss": 6.4437,
"mean_token_accuracy": 0.1176496423780918,
"num_tokens": 1797975.0,
"step": 1035
},
{
"entropy": 6.498772668838501,
"epoch": 0.08091810931725345,
"grad_norm": 1.125,
"learning_rate": 0.0004999998961456145,
"loss": 6.5537,
"mean_token_accuracy": 0.11998764723539353,
"num_tokens": 1806129.0,
"step": 1040
},
{
"entropy": 6.631723117828369,
"epoch": 0.08130713868897102,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999998678092794,
"loss": 6.5327,
"mean_token_accuracy": 0.11050191894173622,
"num_tokens": 1813956.0,
"step": 1045
},
{
"entropy": 6.599989175796509,
"epoch": 0.08169616806068858,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999998360589293,
"loss": 6.5657,
"mean_token_accuracy": 0.11095986440777779,
"num_tokens": 1822572.0,
"step": 1050
},
{
"entropy": 6.480882072448731,
"epoch": 0.08208519743240615,
"grad_norm": 1.0703125,
"learning_rate": 0.000499999800894565,
"loss": 6.4681,
"mean_token_accuracy": 0.11177131086587906,
"num_tokens": 1831870.0,
"step": 1055
},
{
"entropy": 6.47617883682251,
"epoch": 0.08247422680412371,
"grad_norm": 1.1875,
"learning_rate": 0.0004999997623161866,
"loss": 6.5107,
"mean_token_accuracy": 0.1112887792289257,
"num_tokens": 1840414.0,
"step": 1060
},
{
"entropy": 6.575576829910278,
"epoch": 0.08286325617584128,
"grad_norm": 1.078125,
"learning_rate": 0.0004999997203237951,
"loss": 6.527,
"mean_token_accuracy": 0.1100434735417366,
"num_tokens": 1850385.0,
"step": 1065
},
{
"entropy": 6.428466129302978,
"epoch": 0.08325228554755884,
"grad_norm": 1.125,
"learning_rate": 0.000499999674917391,
"loss": 6.4336,
"mean_token_accuracy": 0.12001706510782242,
"num_tokens": 1858479.0,
"step": 1070
},
{
"entropy": 6.507694816589355,
"epoch": 0.0836413149192764,
"grad_norm": 1.046875,
"learning_rate": 0.0004999996260969748,
"loss": 6.4857,
"mean_token_accuracy": 0.11208872050046921,
"num_tokens": 1867357.0,
"step": 1075
},
{
"entropy": 6.512378454208374,
"epoch": 0.08403034429099397,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999995738625475,
"loss": 6.4521,
"mean_token_accuracy": 0.1077382005751133,
"num_tokens": 1876211.0,
"step": 1080
},
{
"entropy": 6.390499639511108,
"epoch": 0.08441937366271153,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999995182141099,
"loss": 6.3633,
"mean_token_accuracy": 0.1128227360546589,
"num_tokens": 1884301.0,
"step": 1085
},
{
"entropy": 6.635722494125366,
"epoch": 0.0848084030344291,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999994591516626,
"loss": 6.5416,
"mean_token_accuracy": 0.11540148556232452,
"num_tokens": 1893191.0,
"step": 1090
},
{
"entropy": 6.37746410369873,
"epoch": 0.08519743240614666,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999993966752066,
"loss": 6.4131,
"mean_token_accuracy": 0.11306635513901711,
"num_tokens": 1901434.0,
"step": 1095
},
{
"entropy": 6.484576797485351,
"epoch": 0.08558646177786423,
"grad_norm": 1.140625,
"learning_rate": 0.000499999330784743,
"loss": 6.3908,
"mean_token_accuracy": 0.12002351284027099,
"num_tokens": 1910054.0,
"step": 1100
},
{
"entropy": 6.490119457244873,
"epoch": 0.08597549114958179,
"grad_norm": 1.109375,
"learning_rate": 0.0004999992614802725,
"loss": 6.5397,
"mean_token_accuracy": 0.11540758833289147,
"num_tokens": 1918607.0,
"step": 1105
},
{
"entropy": 6.551545763015747,
"epoch": 0.08636452052129936,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999991887617966,
"loss": 6.4828,
"mean_token_accuracy": 0.11358762234449386,
"num_tokens": 1927610.0,
"step": 1110
},
{
"entropy": 6.389538955688477,
"epoch": 0.08675354989301692,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999991126293158,
"loss": 6.4366,
"mean_token_accuracy": 0.11365601867437362,
"num_tokens": 1936174.0,
"step": 1115
},
{
"entropy": 6.459973239898682,
"epoch": 0.08714257926473448,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999990330828318,
"loss": 6.3439,
"mean_token_accuracy": 0.12333748191595077,
"num_tokens": 1944228.0,
"step": 1120
},
{
"entropy": 6.483522081375122,
"epoch": 0.08753160863645205,
"grad_norm": 1.09375,
"learning_rate": 0.0004999989501223456,
"loss": 6.4502,
"mean_token_accuracy": 0.11555779352784157,
"num_tokens": 1953144.0,
"step": 1125
},
{
"entropy": 6.46024751663208,
"epoch": 0.08792063800816961,
"grad_norm": 1.15625,
"learning_rate": 0.0004999988637478584,
"loss": 6.5161,
"mean_token_accuracy": 0.1126246102154255,
"num_tokens": 1961947.0,
"step": 1130
},
{
"entropy": 6.5634512424469,
"epoch": 0.08830966737988719,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999987739593716,
"loss": 6.4809,
"mean_token_accuracy": 0.11571738943457603,
"num_tokens": 1970674.0,
"step": 1135
},
{
"entropy": 6.490915060043335,
"epoch": 0.08869869675160474,
"grad_norm": 1.09375,
"learning_rate": 0.0004999986807568865,
"loss": 6.4496,
"mean_token_accuracy": 0.11765117347240447,
"num_tokens": 1979173.0,
"step": 1140
},
{
"entropy": 6.474772596359253,
"epoch": 0.08908772612332232,
"grad_norm": 1.171875,
"learning_rate": 0.0004999985841404045,
"loss": 6.4392,
"mean_token_accuracy": 0.11643676534295082,
"num_tokens": 1988134.0,
"step": 1145
},
{
"entropy": 6.441124773025512,
"epoch": 0.08947675549503988,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999984841099271,
"loss": 6.3768,
"mean_token_accuracy": 0.12014681100845337,
"num_tokens": 1996977.0,
"step": 1150
},
{
"entropy": 6.50206184387207,
"epoch": 0.08986578486675743,
"grad_norm": 1.0859375,
"learning_rate": 0.000499998380665456,
"loss": 6.4588,
"mean_token_accuracy": 0.1176146350800991,
"num_tokens": 2005946.0,
"step": 1155
},
{
"entropy": 6.388955450057983,
"epoch": 0.090254814238475,
"grad_norm": 1.2890625,
"learning_rate": 0.0004999982738069924,
"loss": 6.2831,
"mean_token_accuracy": 0.12113617286086083,
"num_tokens": 2014882.0,
"step": 1160
},
{
"entropy": 6.391384267807007,
"epoch": 0.09064384361019256,
"grad_norm": 1.203125,
"learning_rate": 0.0004999981635345382,
"loss": 6.4771,
"mean_token_accuracy": 0.1094152309000492,
"num_tokens": 2024858.0,
"step": 1165
},
{
"entropy": 6.371677255630493,
"epoch": 0.09103287298191014,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999980498480949,
"loss": 6.3701,
"mean_token_accuracy": 0.11717021316289902,
"num_tokens": 2034072.0,
"step": 1170
},
{
"entropy": 6.45729169845581,
"epoch": 0.0914219023536277,
"grad_norm": 1.078125,
"learning_rate": 0.0004999979327476644,
"loss": 6.3869,
"mean_token_accuracy": 0.11678455844521522,
"num_tokens": 2042649.0,
"step": 1175
},
{
"entropy": 6.433530855178833,
"epoch": 0.09181093172534527,
"grad_norm": 1.109375,
"learning_rate": 0.0004999978122332485,
"loss": 6.4431,
"mean_token_accuracy": 0.1181255742907524,
"num_tokens": 2050866.0,
"step": 1180
},
{
"entropy": 6.4381969451904295,
"epoch": 0.09219996109706283,
"grad_norm": 1.203125,
"learning_rate": 0.0004999976883048487,
"loss": 6.4242,
"mean_token_accuracy": 0.11656152158975601,
"num_tokens": 2058914.0,
"step": 1185
},
{
"entropy": 6.419391775131226,
"epoch": 0.0925889904687804,
"grad_norm": 1.125,
"learning_rate": 0.0004999975609624673,
"loss": 6.362,
"mean_token_accuracy": 0.11579015702009202,
"num_tokens": 2068098.0,
"step": 1190
},
{
"entropy": 6.418846511840821,
"epoch": 0.09297801984049796,
"grad_norm": 1.078125,
"learning_rate": 0.0004999974302061059,
"loss": 6.3645,
"mean_token_accuracy": 0.11124148517847061,
"num_tokens": 2077299.0,
"step": 1195
},
{
"entropy": 6.40845251083374,
"epoch": 0.09336704921221552,
"grad_norm": 1.25,
"learning_rate": 0.0004999972960357666,
"loss": 6.4689,
"mean_token_accuracy": 0.11288349330425262,
"num_tokens": 2087868.0,
"step": 1200
},
{
"entropy": 6.447312068939209,
"epoch": 0.09375607858393309,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999971584514516,
"loss": 6.3783,
"mean_token_accuracy": 0.123850879073143,
"num_tokens": 2096553.0,
"step": 1205
},
{
"entropy": 6.458125257492066,
"epoch": 0.09414510795565065,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999970174531627,
"loss": 6.3609,
"mean_token_accuracy": 0.11663044691085815,
"num_tokens": 2104809.0,
"step": 1210
},
{
"entropy": 6.294647073745727,
"epoch": 0.09453413732736822,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999968730409022,
"loss": 6.3464,
"mean_token_accuracy": 0.11563503369688988,
"num_tokens": 2113869.0,
"step": 1215
},
{
"entropy": 6.34133825302124,
"epoch": 0.09492316669908578,
"grad_norm": 1.140625,
"learning_rate": 0.0004999967252146723,
"loss": 6.1603,
"mean_token_accuracy": 0.11609101369976997,
"num_tokens": 2121749.0,
"step": 1220
},
{
"entropy": 6.341687250137329,
"epoch": 0.09531219607080335,
"grad_norm": 1.09375,
"learning_rate": 0.0004999965739744752,
"loss": 6.3824,
"mean_token_accuracy": 0.11490623876452447,
"num_tokens": 2130828.0,
"step": 1225
},
{
"entropy": 6.462488079071045,
"epoch": 0.09570122544252091,
"grad_norm": 1.125,
"learning_rate": 0.0004999964193203132,
"loss": 6.35,
"mean_token_accuracy": 0.11846289187669753,
"num_tokens": 2138928.0,
"step": 1230
},
{
"entropy": 6.407934045791626,
"epoch": 0.09609025481423847,
"grad_norm": 1.0625,
"learning_rate": 0.0004999962612521888,
"loss": 6.3864,
"mean_token_accuracy": 0.11911074891686439,
"num_tokens": 2147643.0,
"step": 1235
},
{
"entropy": 6.398451328277588,
"epoch": 0.09647928418595604,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999960997701041,
"loss": 6.3112,
"mean_token_accuracy": 0.12286981865763665,
"num_tokens": 2156766.0,
"step": 1240
},
{
"entropy": 6.415870094299317,
"epoch": 0.0968683135576736,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999959348740618,
"loss": 6.3735,
"mean_token_accuracy": 0.11572954803705215,
"num_tokens": 2165517.0,
"step": 1245
},
{
"entropy": 6.276720762252808,
"epoch": 0.09725734292939117,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999957665640641,
"loss": 6.2391,
"mean_token_accuracy": 0.12221750989556313,
"num_tokens": 2174812.0,
"step": 1250
},
{
"entropy": 6.389494466781616,
"epoch": 0.09764637230110873,
"grad_norm": 1.046875,
"learning_rate": 0.000499995594840114,
"loss": 6.3961,
"mean_token_accuracy": 0.1142569437623024,
"num_tokens": 2183968.0,
"step": 1255
},
{
"entropy": 6.395362615585327,
"epoch": 0.0980354016728263,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999954197022138,
"loss": 6.4207,
"mean_token_accuracy": 0.12095934301614761,
"num_tokens": 2191947.0,
"step": 1260
},
{
"entropy": 6.362047100067139,
"epoch": 0.09842443104454386,
"grad_norm": 1.09375,
"learning_rate": 0.0004999952411503661,
"loss": 6.2524,
"mean_token_accuracy": 0.12708474695682526,
"num_tokens": 2199951.0,
"step": 1265
},
{
"entropy": 6.345381879806519,
"epoch": 0.09881346041626143,
"grad_norm": 1.1875,
"learning_rate": 0.0004999950591845738,
"loss": 6.4025,
"mean_token_accuracy": 0.11825087368488311,
"num_tokens": 2208714.0,
"step": 1270
},
{
"entropy": 6.296407985687256,
"epoch": 0.09920248978797899,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999948738048397,
"loss": 6.2605,
"mean_token_accuracy": 0.12081179469823837,
"num_tokens": 2217250.0,
"step": 1275
},
{
"entropy": 6.371564292907715,
"epoch": 0.09959151915969655,
"grad_norm": 1.203125,
"learning_rate": 0.0004999946850111663,
"loss": 6.329,
"mean_token_accuracy": 0.12220223546028137,
"num_tokens": 2225468.0,
"step": 1280
},
{
"entropy": 6.248642206192017,
"epoch": 0.09998054853141412,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999944928035569,
"loss": 6.2729,
"mean_token_accuracy": 0.12882675752043724,
"num_tokens": 2233893.0,
"step": 1285
},
{
"entropy": 6.485666704177857,
"epoch": 0.10036957790313168,
"grad_norm": 1.15625,
"learning_rate": 0.000499994297182014,
"loss": 6.3387,
"mean_token_accuracy": 0.11729943826794624,
"num_tokens": 2242487.0,
"step": 1290
},
{
"entropy": 6.232321691513062,
"epoch": 0.10075860727484925,
"grad_norm": 1.109375,
"learning_rate": 0.0004999940981465408,
"loss": 6.1769,
"mean_token_accuracy": 0.12584130242466926,
"num_tokens": 2250518.0,
"step": 1295
},
{
"entropy": 6.359662485122681,
"epoch": 0.10114763664656681,
"grad_norm": 1.15625,
"learning_rate": 0.0004999938956971404,
"loss": 6.35,
"mean_token_accuracy": 0.11376101300120353,
"num_tokens": 2259024.0,
"step": 1300
},
{
"entropy": 6.315293645858764,
"epoch": 0.10153666601828439,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999936898338156,
"loss": 6.406,
"mean_token_accuracy": 0.11127450317144394,
"num_tokens": 2267832.0,
"step": 1305
},
{
"entropy": 6.441192197799682,
"epoch": 0.10192569539000194,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999934805565698,
"loss": 6.4422,
"mean_token_accuracy": 0.11080272123217583,
"num_tokens": 2276834.0,
"step": 1310
},
{
"entropy": 6.482887125015258,
"epoch": 0.10231472476171952,
"grad_norm": 1.0625,
"learning_rate": 0.000499993267865406,
"loss": 6.3746,
"mean_token_accuracy": 0.11520213186740876,
"num_tokens": 2285736.0,
"step": 1315
},
{
"entropy": 6.384394311904908,
"epoch": 0.10270375413343708,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999930517603275,
"loss": 6.3234,
"mean_token_accuracy": 0.1279608964920044,
"num_tokens": 2294177.0,
"step": 1320
},
{
"entropy": 6.272417211532593,
"epoch": 0.10309278350515463,
"grad_norm": 1.171875,
"learning_rate": 0.0004999928322413374,
"loss": 6.1152,
"mean_token_accuracy": 0.13164947628974916,
"num_tokens": 2302784.0,
"step": 1325
},
{
"entropy": 6.1753613471984865,
"epoch": 0.1034818128768722,
"grad_norm": 1.0625,
"learning_rate": 0.0004999926093084394,
"loss": 6.2247,
"mean_token_accuracy": 0.1209910273551941,
"num_tokens": 2311650.0,
"step": 1330
},
{
"entropy": 6.426689910888672,
"epoch": 0.10387084224858976,
"grad_norm": 1.09375,
"learning_rate": 0.0004999923829616365,
"loss": 6.3461,
"mean_token_accuracy": 0.12103552594780922,
"num_tokens": 2319828.0,
"step": 1335
},
{
"entropy": 6.360848712921142,
"epoch": 0.10425987162030734,
"grad_norm": 1.109375,
"learning_rate": 0.0004999921532009323,
"loss": 6.3423,
"mean_token_accuracy": 0.11625662669539452,
"num_tokens": 2328636.0,
"step": 1340
},
{
"entropy": 6.244861602783203,
"epoch": 0.1046489009920249,
"grad_norm": 1.1875,
"learning_rate": 0.0004999919200263304,
"loss": 6.2352,
"mean_token_accuracy": 0.1234541155397892,
"num_tokens": 2336771.0,
"step": 1345
},
{
"entropy": 6.352584886550903,
"epoch": 0.10503793036374247,
"grad_norm": 1.046875,
"learning_rate": 0.0004999916834378342,
"loss": 6.3854,
"mean_token_accuracy": 0.11652181446552276,
"num_tokens": 2346106.0,
"step": 1350
},
{
"entropy": 6.42938871383667,
"epoch": 0.10542695973546003,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999914434354472,
"loss": 6.3661,
"mean_token_accuracy": 0.11678783521056176,
"num_tokens": 2354943.0,
"step": 1355
},
{
"entropy": 6.208476591110229,
"epoch": 0.10581598910717759,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999912000191733,
"loss": 6.2707,
"mean_token_accuracy": 0.12307048067450524,
"num_tokens": 2363699.0,
"step": 1360
},
{
"entropy": 6.340670251846314,
"epoch": 0.10620501847889516,
"grad_norm": 1.1015625,
"learning_rate": 0.000499990953189016,
"loss": 6.4154,
"mean_token_accuracy": 0.11648116931319237,
"num_tokens": 2373316.0,
"step": 1365
},
{
"entropy": 6.536358499526978,
"epoch": 0.10659404785061272,
"grad_norm": 1.140625,
"learning_rate": 0.0004999907029449791,
"loss": 6.4774,
"mean_token_accuracy": 0.1135114923119545,
"num_tokens": 2381550.0,
"step": 1370
},
{
"entropy": 6.30794472694397,
"epoch": 0.10698307722233029,
"grad_norm": 1.140625,
"learning_rate": 0.0004999904492870663,
"loss": 6.2883,
"mean_token_accuracy": 0.12282988801598549,
"num_tokens": 2389546.0,
"step": 1375
},
{
"entropy": 6.380693864822388,
"epoch": 0.10737210659404785,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999901922152816,
"loss": 6.244,
"mean_token_accuracy": 0.12719449624419213,
"num_tokens": 2398852.0,
"step": 1380
},
{
"entropy": 6.272253751754761,
"epoch": 0.10776113596576542,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999899317296288,
"loss": 6.3308,
"mean_token_accuracy": 0.12012203708291054,
"num_tokens": 2407815.0,
"step": 1385
},
{
"entropy": 6.317573165893554,
"epoch": 0.10815016533748298,
"grad_norm": 1.046875,
"learning_rate": 0.0004999896678301121,
"loss": 6.1922,
"mean_token_accuracy": 0.12548744305968285,
"num_tokens": 2417300.0,
"step": 1390
},
{
"entropy": 6.297150897979736,
"epoch": 0.10853919470920055,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999894005167352,
"loss": 6.3578,
"mean_token_accuracy": 0.11943556964397431,
"num_tokens": 2426257.0,
"step": 1395
},
{
"entropy": 6.4665430068969725,
"epoch": 0.10892822408091811,
"grad_norm": 1.0703125,
"learning_rate": 0.000499989129789502,
"loss": 6.3817,
"mean_token_accuracy": 0.12104496955871583,
"num_tokens": 2434595.0,
"step": 1400
},
{
"entropy": 6.406263446807861,
"epoch": 0.10931725345263567,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999888556484172,
"loss": 6.39,
"mean_token_accuracy": 0.11528819128870964,
"num_tokens": 2443945.0,
"step": 1405
},
{
"entropy": 6.226689291000366,
"epoch": 0.10970628282435324,
"grad_norm": 1.125,
"learning_rate": 0.0004999885780934845,
"loss": 6.2536,
"mean_token_accuracy": 0.12194890603423118,
"num_tokens": 2453141.0,
"step": 1410
},
{
"entropy": 6.352630996704102,
"epoch": 0.1100953121960708,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999882971247081,
"loss": 6.3336,
"mean_token_accuracy": 0.122190822660923,
"num_tokens": 2461873.0,
"step": 1415
},
{
"entropy": 6.317797040939331,
"epoch": 0.11048434156778837,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999880127420926,
"loss": 6.3632,
"mean_token_accuracy": 0.11227262541651725,
"num_tokens": 2471238.0,
"step": 1420
},
{
"entropy": 6.334433937072754,
"epoch": 0.11087337093950593,
"grad_norm": 1.203125,
"learning_rate": 0.0004999877249456421,
"loss": 6.2408,
"mean_token_accuracy": 0.1272654689848423,
"num_tokens": 2479412.0,
"step": 1425
},
{
"entropy": 6.188841533660889,
"epoch": 0.1112624003112235,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999874337353609,
"loss": 6.1292,
"mean_token_accuracy": 0.12458490431308747,
"num_tokens": 2488150.0,
"step": 1430
},
{
"entropy": 6.280575132369995,
"epoch": 0.11165142968294106,
"grad_norm": 1.046875,
"learning_rate": 0.0004999871391112535,
"loss": 6.309,
"mean_token_accuracy": 0.12516588494181632,
"num_tokens": 2498103.0,
"step": 1435
},
{
"entropy": 6.1986839294433596,
"epoch": 0.11204045905465862,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999868410733244,
"loss": 6.1695,
"mean_token_accuracy": 0.125221885740757,
"num_tokens": 2506211.0,
"step": 1440
},
{
"entropy": 6.172796535491943,
"epoch": 0.11242948842637619,
"grad_norm": 1.140625,
"learning_rate": 0.000499986539621578,
"loss": 6.2162,
"mean_token_accuracy": 0.12384542673826218,
"num_tokens": 2514743.0,
"step": 1445
},
{
"entropy": 6.167073822021484,
"epoch": 0.11281851779809375,
"grad_norm": 1.03125,
"learning_rate": 0.0004999862347560191,
"loss": 6.2027,
"mean_token_accuracy": 0.12268847003579139,
"num_tokens": 2523824.0,
"step": 1450
},
{
"entropy": 6.387971639633179,
"epoch": 0.11320754716981132,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999859264766521,
"loss": 6.2719,
"mean_token_accuracy": 0.12253661304712296,
"num_tokens": 2532360.0,
"step": 1455
},
{
"entropy": 6.257190561294555,
"epoch": 0.11359657654152888,
"grad_norm": 1.109375,
"learning_rate": 0.0004999856147834817,
"loss": 6.2217,
"mean_token_accuracy": 0.1282711908221245,
"num_tokens": 2540547.0,
"step": 1460
},
{
"entropy": 6.174588632583618,
"epoch": 0.11398560591324645,
"grad_norm": 1.140625,
"learning_rate": 0.0004999852996765129,
"loss": 6.2459,
"mean_token_accuracy": 0.12327087596058846,
"num_tokens": 2549193.0,
"step": 1465
},
{
"entropy": 6.317156839370727,
"epoch": 0.11437463528496401,
"grad_norm": 1.09375,
"learning_rate": 0.0004999849811557502,
"loss": 6.254,
"mean_token_accuracy": 0.12232557386159897,
"num_tokens": 2557553.0,
"step": 1470
},
{
"entropy": 6.374777555465698,
"epoch": 0.11476366465668159,
"grad_norm": 1.09375,
"learning_rate": 0.0004999846592211984,
"loss": 6.3602,
"mean_token_accuracy": 0.11835889741778374,
"num_tokens": 2565945.0,
"step": 1475
},
{
"entropy": 6.324231147766113,
"epoch": 0.11515269402839914,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999843338728626,
"loss": 6.2285,
"mean_token_accuracy": 0.11772384941577911,
"num_tokens": 2574364.0,
"step": 1480
},
{
"entropy": 6.176779270172119,
"epoch": 0.1155417234001167,
"grad_norm": 1.0,
"learning_rate": 0.0004999840051107476,
"loss": 6.1531,
"mean_token_accuracy": 0.1252761095762253,
"num_tokens": 2582825.0,
"step": 1485
},
{
"entropy": 6.367834663391113,
"epoch": 0.11593075277183428,
"grad_norm": 1.078125,
"learning_rate": 0.0004999836729348584,
"loss": 6.2943,
"mean_token_accuracy": 0.12382682561874389,
"num_tokens": 2591628.0,
"step": 1490
},
{
"entropy": 6.331791353225708,
"epoch": 0.11631978214355183,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999833373452,
"loss": 6.2892,
"mean_token_accuracy": 0.1190582662820816,
"num_tokens": 2600436.0,
"step": 1495
},
{
"entropy": 6.217423963546753,
"epoch": 0.1167088115152694,
"grad_norm": 1.1875,
"learning_rate": 0.0004999829983417778,
"loss": 6.2596,
"mean_token_accuracy": 0.12188921868801117,
"num_tokens": 2609709.0,
"step": 1500
},
{
"entropy": 6.359324789047241,
"epoch": 0.11709784088698696,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999826559245965,
"loss": 6.2586,
"mean_token_accuracy": 0.12367889285087585,
"num_tokens": 2617992.0,
"step": 1505
},
{
"entropy": 6.17459282875061,
"epoch": 0.11748687025870454,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999823100936615,
"loss": 6.1695,
"mean_token_accuracy": 0.12516930997371672,
"num_tokens": 2626706.0,
"step": 1510
},
{
"entropy": 6.3931690692901615,
"epoch": 0.1178758996304221,
"grad_norm": 1.203125,
"learning_rate": 0.0004999819608489781,
"loss": 6.2359,
"mean_token_accuracy": 0.13169281259179116,
"num_tokens": 2634642.0,
"step": 1515
},
{
"entropy": 6.271792840957642,
"epoch": 0.11826492900213967,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999816081905515,
"loss": 6.2722,
"mean_token_accuracy": 0.11658980697393417,
"num_tokens": 2644160.0,
"step": 1520
},
{
"entropy": 6.214734745025635,
"epoch": 0.11865395837385723,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999812521183872,
"loss": 6.2285,
"mean_token_accuracy": 0.1217222385108471,
"num_tokens": 2653390.0,
"step": 1525
},
{
"entropy": 6.356004905700684,
"epoch": 0.11904298774557479,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999808926324903,
"loss": 6.2785,
"mean_token_accuracy": 0.12192253395915031,
"num_tokens": 2662561.0,
"step": 1530
},
{
"entropy": 6.280083322525025,
"epoch": 0.11943201711729236,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999805297328665,
"loss": 6.2453,
"mean_token_accuracy": 0.12399770319461823,
"num_tokens": 2670694.0,
"step": 1535
},
{
"entropy": 6.214053916931152,
"epoch": 0.11982104648900992,
"grad_norm": 1.125,
"learning_rate": 0.0004999801634195214,
"loss": 6.1408,
"mean_token_accuracy": 0.12651772275567055,
"num_tokens": 2679583.0,
"step": 1540
},
{
"entropy": 6.181328582763672,
"epoch": 0.12021007586072749,
"grad_norm": 1.09375,
"learning_rate": 0.0004999797936924603,
"loss": 6.1127,
"mean_token_accuracy": 0.12723355293273925,
"num_tokens": 2688429.0,
"step": 1545
},
{
"entropy": 6.365040397644043,
"epoch": 0.12059910523244505,
"grad_norm": 1.109375,
"learning_rate": 0.000499979420551689,
"loss": 6.2115,
"mean_token_accuracy": 0.12122456356883049,
"num_tokens": 2697134.0,
"step": 1550
},
{
"entropy": 6.182654905319214,
"epoch": 0.12098813460416262,
"grad_norm": 1.09375,
"learning_rate": 0.000499979043997213,
"loss": 6.3483,
"mean_token_accuracy": 0.11888098418712616,
"num_tokens": 2705641.0,
"step": 1555
},
{
"entropy": 6.314460563659668,
"epoch": 0.12137716397588018,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999786640290381,
"loss": 6.2908,
"mean_token_accuracy": 0.12467942461371422,
"num_tokens": 2714977.0,
"step": 1560
},
{
"entropy": 6.2193540096282955,
"epoch": 0.12176619334759774,
"grad_norm": 1.1875,
"learning_rate": 0.0004999782806471701,
"loss": 6.1337,
"mean_token_accuracy": 0.1347135454416275,
"num_tokens": 2723120.0,
"step": 1565
},
{
"entropy": 6.0879443168640135,
"epoch": 0.12215522271931531,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999778938516147,
"loss": 6.1317,
"mean_token_accuracy": 0.13238951042294503,
"num_tokens": 2731964.0,
"step": 1570
},
{
"entropy": 6.39334282875061,
"epoch": 0.12254425209103287,
"grad_norm": 1.0703125,
"learning_rate": 0.000499977503642378,
"loss": 6.1931,
"mean_token_accuracy": 0.127005735039711,
"num_tokens": 2741537.0,
"step": 1575
},
{
"entropy": 6.138741397857666,
"epoch": 0.12293328146275044,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999771100194656,
"loss": 6.3277,
"mean_token_accuracy": 0.1212069720029831,
"num_tokens": 2751522.0,
"step": 1580
},
{
"entropy": 6.3773661136627195,
"epoch": 0.123322310834468,
"grad_norm": 1.03125,
"learning_rate": 0.0004999767129828837,
"loss": 6.3125,
"mean_token_accuracy": 0.11907947659492493,
"num_tokens": 2761100.0,
"step": 1585
},
{
"entropy": 6.22048397064209,
"epoch": 0.12371134020618557,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999763125326383,
"loss": 6.2478,
"mean_token_accuracy": 0.12256922721862792,
"num_tokens": 2769171.0,
"step": 1590
},
{
"entropy": 6.243303442001343,
"epoch": 0.12410036957790313,
"grad_norm": 1.140625,
"learning_rate": 0.0004999759086687355,
"loss": 6.1867,
"mean_token_accuracy": 0.12792484536767007,
"num_tokens": 2778021.0,
"step": 1595
},
{
"entropy": 6.166972827911377,
"epoch": 0.1244893989496207,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999755013911813,
"loss": 6.1195,
"mean_token_accuracy": 0.13010439202189444,
"num_tokens": 2787206.0,
"step": 1600
},
{
"entropy": 6.148937082290649,
"epoch": 0.12487842832133826,
"grad_norm": 1.09375,
"learning_rate": 0.000499975090699982,
"loss": 6.1981,
"mean_token_accuracy": 0.1204142227768898,
"num_tokens": 2796051.0,
"step": 1605
},
{
"entropy": 6.247499084472656,
"epoch": 0.12526745769305583,
"grad_norm": 1.015625,
"learning_rate": 0.0004999746765951438,
"loss": 6.144,
"mean_token_accuracy": 0.12628112062811853,
"num_tokens": 2804961.0,
"step": 1610
},
{
"entropy": 6.156633281707764,
"epoch": 0.1256564870647734,
"grad_norm": 1.140625,
"learning_rate": 0.0004999742590766729,
"loss": 6.0863,
"mean_token_accuracy": 0.13187484741210936,
"num_tokens": 2813691.0,
"step": 1615
},
{
"entropy": 6.225268125534058,
"epoch": 0.12604551643649095,
"grad_norm": 1.140625,
"learning_rate": 0.0004999738381445757,
"loss": 6.248,
"mean_token_accuracy": 0.12675216943025588,
"num_tokens": 2822248.0,
"step": 1620
},
{
"entropy": 6.251285171508789,
"epoch": 0.1264345458082085,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999734137988586,
"loss": 6.1923,
"mean_token_accuracy": 0.12956587076187134,
"num_tokens": 2830826.0,
"step": 1625
},
{
"entropy": 6.216675090789795,
"epoch": 0.1268235751799261,
"grad_norm": 1.1875,
"learning_rate": 0.000499972986039528,
"loss": 6.1839,
"mean_token_accuracy": 0.12593767046928406,
"num_tokens": 2839766.0,
"step": 1630
},
{
"entropy": 6.123473691940307,
"epoch": 0.12721260455164365,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999725548665904,
"loss": 6.1228,
"mean_token_accuracy": 0.13010624870657922,
"num_tokens": 2848728.0,
"step": 1635
},
{
"entropy": 6.1722784519195555,
"epoch": 0.1276016339233612,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999721202800524,
"loss": 6.1396,
"mean_token_accuracy": 0.13079047277569772,
"num_tokens": 2857221.0,
"step": 1640
},
{
"entropy": 6.263162755966187,
"epoch": 0.12799066329507877,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999716822799206,
"loss": 6.2131,
"mean_token_accuracy": 0.12727918699383736,
"num_tokens": 2865314.0,
"step": 1645
},
{
"entropy": 6.1169415473937985,
"epoch": 0.12837969266679633,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999712408662016,
"loss": 6.088,
"mean_token_accuracy": 0.12827678620815278,
"num_tokens": 2873654.0,
"step": 1650
},
{
"entropy": 6.114794111251831,
"epoch": 0.12876872203851392,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999707960389021,
"loss": 5.9976,
"mean_token_accuracy": 0.1390816293656826,
"num_tokens": 2881604.0,
"step": 1655
},
{
"entropy": 6.175813865661621,
"epoch": 0.12915775141023147,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999703477980288,
"loss": 6.1899,
"mean_token_accuracy": 0.12590051293373108,
"num_tokens": 2890427.0,
"step": 1660
},
{
"entropy": 6.176800918579102,
"epoch": 0.12954678078194903,
"grad_norm": 1.15625,
"learning_rate": 0.0004999698961435885,
"loss": 6.1608,
"mean_token_accuracy": 0.1278068646788597,
"num_tokens": 2899206.0,
"step": 1665
},
{
"entropy": 6.1610307693481445,
"epoch": 0.1299358101536666,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999694410755882,
"loss": 6.0803,
"mean_token_accuracy": 0.1307940810918808,
"num_tokens": 2907738.0,
"step": 1670
},
{
"entropy": 6.264579439163208,
"epoch": 0.13032483952538418,
"grad_norm": 1.0625,
"learning_rate": 0.0004999689825940347,
"loss": 6.3124,
"mean_token_accuracy": 0.12490758523344994,
"num_tokens": 2916881.0,
"step": 1675
},
{
"entropy": 6.203776741027832,
"epoch": 0.13071386889710174,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999685206989349,
"loss": 6.2506,
"mean_token_accuracy": 0.12321256548166275,
"num_tokens": 2926467.0,
"step": 1680
},
{
"entropy": 6.266940355300903,
"epoch": 0.1311028982688193,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999680553902959,
"loss": 6.1586,
"mean_token_accuracy": 0.12849143967032434,
"num_tokens": 2934671.0,
"step": 1685
},
{
"entropy": 6.153891229629517,
"epoch": 0.13149192764053685,
"grad_norm": 1.015625,
"learning_rate": 0.0004999675866681247,
"loss": 6.1937,
"mean_token_accuracy": 0.12307270169258118,
"num_tokens": 2943925.0,
"step": 1690
},
{
"entropy": 6.251479959487915,
"epoch": 0.1318809570122544,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999671145324286,
"loss": 6.2351,
"mean_token_accuracy": 0.12551012486219407,
"num_tokens": 2952118.0,
"step": 1695
},
{
"entropy": 6.142472887039185,
"epoch": 0.132269986383972,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999666389832144,
"loss": 6.2185,
"mean_token_accuracy": 0.12478951662778855,
"num_tokens": 2961110.0,
"step": 1700
},
{
"entropy": 6.258459663391113,
"epoch": 0.13265901575568956,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999661600204897,
"loss": 6.1492,
"mean_token_accuracy": 0.1310803025960922,
"num_tokens": 2968877.0,
"step": 1705
},
{
"entropy": 6.205829286575318,
"epoch": 0.13304804512740712,
"grad_norm": 1.09375,
"learning_rate": 0.0004999656776442615,
"loss": 6.2062,
"mean_token_accuracy": 0.12969572320580483,
"num_tokens": 2977346.0,
"step": 1710
},
{
"entropy": 6.306067752838135,
"epoch": 0.13343707449912467,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999651918545372,
"loss": 6.2403,
"mean_token_accuracy": 0.12232028320431709,
"num_tokens": 2987494.0,
"step": 1715
},
{
"entropy": 6.168236112594604,
"epoch": 0.13382610387084226,
"grad_norm": 1.09375,
"learning_rate": 0.0004999647026513243,
"loss": 6.167,
"mean_token_accuracy": 0.12709078639745713,
"num_tokens": 2996728.0,
"step": 1720
},
{
"entropy": 6.16880464553833,
"epoch": 0.13421513324255982,
"grad_norm": 1.140625,
"learning_rate": 0.00049996421003463,
"loss": 6.1947,
"mean_token_accuracy": 0.12163503617048263,
"num_tokens": 3005810.0,
"step": 1725
},
{
"entropy": 6.225845766067505,
"epoch": 0.13460416261427738,
"grad_norm": 1.171875,
"learning_rate": 0.0004999637140044619,
"loss": 6.2057,
"mean_token_accuracy": 0.12583187147974967,
"num_tokens": 3014603.0,
"step": 1730
},
{
"entropy": 6.286410093307495,
"epoch": 0.13499319198599494,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999632145608275,
"loss": 6.2073,
"mean_token_accuracy": 0.12976924255490302,
"num_tokens": 3023170.0,
"step": 1735
},
{
"entropy": 6.047111463546753,
"epoch": 0.1353822213577125,
"grad_norm": 1.140625,
"learning_rate": 0.0004999627117037343,
"loss": 5.952,
"mean_token_accuracy": 0.13934574648737907,
"num_tokens": 3030565.0,
"step": 1740
},
{
"entropy": 6.201064538955689,
"epoch": 0.13577125072943008,
"grad_norm": 1.0234375,
"learning_rate": 0.00049996220543319,
"loss": 6.2296,
"mean_token_accuracy": 0.1313652738928795,
"num_tokens": 3040124.0,
"step": 1745
},
{
"entropy": 6.118889284133911,
"epoch": 0.13616028010114764,
"grad_norm": 1.125,
"learning_rate": 0.0004999616957492025,
"loss": 6.0833,
"mean_token_accuracy": 0.1335503078997135,
"num_tokens": 3048458.0,
"step": 1750
},
{
"entropy": 6.115156173706055,
"epoch": 0.1365493094728652,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999611826517793,
"loss": 6.0529,
"mean_token_accuracy": 0.12220517992973327,
"num_tokens": 3057926.0,
"step": 1755
},
{
"entropy": 6.074193906784058,
"epoch": 0.13693833884458276,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999606661409281,
"loss": 6.0902,
"mean_token_accuracy": 0.12791807651519777,
"num_tokens": 3066683.0,
"step": 1760
},
{
"entropy": 6.162175607681275,
"epoch": 0.13732736821630034,
"grad_norm": 1.09375,
"learning_rate": 0.0004999601462166569,
"loss": 6.1736,
"mean_token_accuracy": 0.12800901383161545,
"num_tokens": 3075117.0,
"step": 1765
},
{
"entropy": 6.291727209091187,
"epoch": 0.1377163975880179,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999596228789736,
"loss": 6.2362,
"mean_token_accuracy": 0.12387871071696281,
"num_tokens": 3084399.0,
"step": 1770
},
{
"entropy": 6.073335409164429,
"epoch": 0.13810542695973546,
"grad_norm": 1.046875,
"learning_rate": 0.0004999590961278859,
"loss": 6.0154,
"mean_token_accuracy": 0.13292096480727195,
"num_tokens": 3093030.0,
"step": 1775
},
{
"entropy": 6.19185700416565,
"epoch": 0.13849445633145302,
"grad_norm": 1.109375,
"learning_rate": 0.0004999585659634021,
"loss": 6.0716,
"mean_token_accuracy": 0.13075146377086638,
"num_tokens": 3101670.0,
"step": 1780
},
{
"entropy": 6.05259747505188,
"epoch": 0.13888348570317058,
"grad_norm": 1.0625,
"learning_rate": 0.0004999580323855301,
"loss": 6.0614,
"mean_token_accuracy": 0.13134398460388183,
"num_tokens": 3109703.0,
"step": 1785
},
{
"entropy": 6.102718019485474,
"epoch": 0.13927251507488816,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999574953942781,
"loss": 6.0242,
"mean_token_accuracy": 0.13451038673520088,
"num_tokens": 3118388.0,
"step": 1790
},
{
"entropy": 6.110586404800415,
"epoch": 0.13966154444660572,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999569549896541,
"loss": 6.0341,
"mean_token_accuracy": 0.13156714141368867,
"num_tokens": 3126466.0,
"step": 1795
},
{
"entropy": 6.112448740005493,
"epoch": 0.14005057381832328,
"grad_norm": 1.0625,
"learning_rate": 0.0004999564111716665,
"loss": 6.0781,
"mean_token_accuracy": 0.1335633009672165,
"num_tokens": 3135322.0,
"step": 1800
},
{
"entropy": 6.108473491668701,
"epoch": 0.14043960319004084,
"grad_norm": 1.078125,
"learning_rate": 0.0004999558639403232,
"loss": 6.0755,
"mean_token_accuracy": 0.13613163232803344,
"num_tokens": 3144498.0,
"step": 1805
},
{
"entropy": 6.045802354812622,
"epoch": 0.1408286325617584,
"grad_norm": 1.0625,
"learning_rate": 0.0004999553132956328,
"loss": 6.0512,
"mean_token_accuracy": 0.12877680659294127,
"num_tokens": 3153573.0,
"step": 1810
},
{
"entropy": 6.046996307373047,
"epoch": 0.14121766193347599,
"grad_norm": 1.078125,
"learning_rate": 0.0004999547592376035,
"loss": 5.9939,
"mean_token_accuracy": 0.1377910055220127,
"num_tokens": 3161705.0,
"step": 1815
},
{
"entropy": 6.138423156738281,
"epoch": 0.14160669130519354,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999542017662438,
"loss": 6.1474,
"mean_token_accuracy": 0.13016104623675345,
"num_tokens": 3170068.0,
"step": 1820
},
{
"entropy": 6.134330701828003,
"epoch": 0.1419957206769111,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999536408815623,
"loss": 6.152,
"mean_token_accuracy": 0.12710127755999565,
"num_tokens": 3178140.0,
"step": 1825
},
{
"entropy": 6.159389972686768,
"epoch": 0.14238475004862866,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999530765835672,
"loss": 6.117,
"mean_token_accuracy": 0.1319376789033413,
"num_tokens": 3186518.0,
"step": 1830
},
{
"entropy": 6.114027595520019,
"epoch": 0.14277377942034625,
"grad_norm": 1.109375,
"learning_rate": 0.0004999525088722673,
"loss": 6.1086,
"mean_token_accuracy": 0.126016952842474,
"num_tokens": 3195246.0,
"step": 1835
},
{
"entropy": 6.2064361572265625,
"epoch": 0.1431628087920638,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999519377476711,
"loss": 6.1239,
"mean_token_accuracy": 0.12959307059645653,
"num_tokens": 3203946.0,
"step": 1840
},
{
"entropy": 6.125613784790039,
"epoch": 0.14355183816378136,
"grad_norm": 1.171875,
"learning_rate": 0.0004999513632097872,
"loss": 6.1511,
"mean_token_accuracy": 0.12730584666132927,
"num_tokens": 3212208.0,
"step": 1845
},
{
"entropy": 6.112041807174682,
"epoch": 0.14394086753549892,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999507852586245,
"loss": 6.078,
"mean_token_accuracy": 0.12968963086605073,
"num_tokens": 3221572.0,
"step": 1850
},
{
"entropy": 6.241564416885376,
"epoch": 0.14432989690721648,
"grad_norm": 1.09375,
"learning_rate": 0.0004999502038941917,
"loss": 6.2186,
"mean_token_accuracy": 0.11879752650856971,
"num_tokens": 3230898.0,
"step": 1855
},
{
"entropy": 6.053813362121582,
"epoch": 0.14471892627893407,
"grad_norm": 1.046875,
"learning_rate": 0.0004999496191164975,
"loss": 6.1755,
"mean_token_accuracy": 0.12418220117688179,
"num_tokens": 3240524.0,
"step": 1860
},
{
"entropy": 6.186307144165039,
"epoch": 0.14510795565065163,
"grad_norm": 1.0703125,
"learning_rate": 0.000499949030925551,
"loss": 5.9339,
"mean_token_accuracy": 0.1323039062321186,
"num_tokens": 3248864.0,
"step": 1865
},
{
"entropy": 6.05433611869812,
"epoch": 0.14549698502236919,
"grad_norm": 1.15625,
"learning_rate": 0.0004999484393213609,
"loss": 6.1267,
"mean_token_accuracy": 0.12983283996582032,
"num_tokens": 3256823.0,
"step": 1870
},
{
"entropy": 6.093509531021118,
"epoch": 0.14588601439408674,
"grad_norm": 1.09375,
"learning_rate": 0.0004999478443039363,
"loss": 6.0317,
"mean_token_accuracy": 0.13464081212878226,
"num_tokens": 3265621.0,
"step": 1875
},
{
"entropy": 6.031926536560059,
"epoch": 0.14627504376580433,
"grad_norm": 1.140625,
"learning_rate": 0.0004999472458732862,
"loss": 6.0619,
"mean_token_accuracy": 0.13131768479943276,
"num_tokens": 3273515.0,
"step": 1880
},
{
"entropy": 6.207943487167358,
"epoch": 0.1466640731375219,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999466440294198,
"loss": 6.131,
"mean_token_accuracy": 0.12989283427596093,
"num_tokens": 3282283.0,
"step": 1885
},
{
"entropy": 6.049416828155517,
"epoch": 0.14705310250923945,
"grad_norm": 1.1328125,
"learning_rate": 0.000499946038772346,
"loss": 6.1341,
"mean_token_accuracy": 0.13318516165018082,
"num_tokens": 3290792.0,
"step": 1890
},
{
"entropy": 6.044079065322876,
"epoch": 0.147442131880957,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999454301020741,
"loss": 5.9918,
"mean_token_accuracy": 0.13286386504769326,
"num_tokens": 3299430.0,
"step": 1895
},
{
"entropy": 6.067660093307495,
"epoch": 0.14783116125267456,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999448180186133,
"loss": 6.0334,
"mean_token_accuracy": 0.13755829334259034,
"num_tokens": 3307949.0,
"step": 1900
},
{
"entropy": 6.091364002227783,
"epoch": 0.14822019062439215,
"grad_norm": 1.109375,
"learning_rate": 0.000499944202521973,
"loss": 6.098,
"mean_token_accuracy": 0.13308693766593932,
"num_tokens": 3316900.0,
"step": 1905
},
{
"entropy": 6.114401006698609,
"epoch": 0.1486092199961097,
"grad_norm": 1.109375,
"learning_rate": 0.0004999435836121624,
"loss": 6.0263,
"mean_token_accuracy": 0.1365790992975235,
"num_tokens": 3325656.0,
"step": 1910
},
{
"entropy": 6.054609870910644,
"epoch": 0.14899824936782727,
"grad_norm": 1.0859375,
"learning_rate": 0.000499942961289191,
"loss": 6.2408,
"mean_token_accuracy": 0.11948857307434083,
"num_tokens": 3334219.0,
"step": 1915
},
{
"entropy": 6.100892877578735,
"epoch": 0.14938727873954483,
"grad_norm": 1.015625,
"learning_rate": 0.0004999423355530683,
"loss": 5.987,
"mean_token_accuracy": 0.1356291465461254,
"num_tokens": 3343637.0,
"step": 1920
},
{
"entropy": 6.047068405151367,
"epoch": 0.1497763081112624,
"grad_norm": 1.09375,
"learning_rate": 0.0004999417064038035,
"loss": 6.1349,
"mean_token_accuracy": 0.1278958112001419,
"num_tokens": 3352624.0,
"step": 1925
},
{
"entropy": 6.089813709259033,
"epoch": 0.15016533748297997,
"grad_norm": 1.0625,
"learning_rate": 0.0004999410738414065,
"loss": 6.0577,
"mean_token_accuracy": 0.13661672845482825,
"num_tokens": 3361394.0,
"step": 1930
},
{
"entropy": 6.1655097007751465,
"epoch": 0.15055436685469753,
"grad_norm": 1.078125,
"learning_rate": 0.0004999404378658866,
"loss": 6.1106,
"mean_token_accuracy": 0.13267300873994828,
"num_tokens": 3369406.0,
"step": 1935
},
{
"entropy": 6.093273782730103,
"epoch": 0.1509433962264151,
"grad_norm": 1.09375,
"learning_rate": 0.0004999397984772536,
"loss": 6.0514,
"mean_token_accuracy": 0.1275075666606426,
"num_tokens": 3377737.0,
"step": 1940
},
{
"entropy": 6.077555751800537,
"epoch": 0.15133242559813265,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999391556755172,
"loss": 6.1175,
"mean_token_accuracy": 0.12663735300302506,
"num_tokens": 3385514.0,
"step": 1945
},
{
"entropy": 6.144791460037231,
"epoch": 0.15172145496985023,
"grad_norm": 1.09375,
"learning_rate": 0.0004999385094606872,
"loss": 6.0261,
"mean_token_accuracy": 0.13351787775754928,
"num_tokens": 3394500.0,
"step": 1950
},
{
"entropy": 6.004178762435913,
"epoch": 0.1521104843415678,
"grad_norm": 1.09375,
"learning_rate": 0.0004999378598327733,
"loss": 6.0057,
"mean_token_accuracy": 0.13412315174937248,
"num_tokens": 3403289.0,
"step": 1955
},
{
"entropy": 6.04636549949646,
"epoch": 0.15249951371328535,
"grad_norm": 1.09375,
"learning_rate": 0.0004999372067917854,
"loss": 5.9441,
"mean_token_accuracy": 0.14245061874389647,
"num_tokens": 3410892.0,
"step": 1960
},
{
"entropy": 5.918894290924072,
"epoch": 0.1528885430850029,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999365503377335,
"loss": 6.0281,
"mean_token_accuracy": 0.13515010178089143,
"num_tokens": 3420329.0,
"step": 1965
},
{
"entropy": 6.066119909286499,
"epoch": 0.1532775724567205,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999358904706273,
"loss": 6.0257,
"mean_token_accuracy": 0.13282209038734435,
"num_tokens": 3429429.0,
"step": 1970
},
{
"entropy": 6.137219715118408,
"epoch": 0.15366660182843805,
"grad_norm": 1.09375,
"learning_rate": 0.0004999352271904771,
"loss": 6.1381,
"mean_token_accuracy": 0.12859977707266806,
"num_tokens": 3438198.0,
"step": 1975
},
{
"entropy": 6.2691608428955075,
"epoch": 0.1540556312001556,
"grad_norm": 1.109375,
"learning_rate": 0.0004999345604972928,
"loss": 6.2676,
"mean_token_accuracy": 0.129219963401556,
"num_tokens": 3447732.0,
"step": 1980
},
{
"entropy": 6.048369979858398,
"epoch": 0.15444466057187317,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999338903910846,
"loss": 6.0545,
"mean_token_accuracy": 0.13279973939061165,
"num_tokens": 3456392.0,
"step": 1985
},
{
"entropy": 6.084347772598266,
"epoch": 0.15483368994359073,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999332168718626,
"loss": 6.0832,
"mean_token_accuracy": 0.1306935466825962,
"num_tokens": 3464473.0,
"step": 1990
},
{
"entropy": 6.114197492599487,
"epoch": 0.15522271931530832,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999325399396371,
"loss": 6.0531,
"mean_token_accuracy": 0.1318880908191204,
"num_tokens": 3473740.0,
"step": 1995
},
{
"entropy": 5.95880823135376,
"epoch": 0.15561174868702587,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999318595944183,
"loss": 5.9461,
"mean_token_accuracy": 0.13854984045028687,
"num_tokens": 3482249.0,
"step": 2000
},
{
"entropy": 6.20720682144165,
"epoch": 0.15600077805874343,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999311758362166,
"loss": 6.194,
"mean_token_accuracy": 0.1284084811806679,
"num_tokens": 3490753.0,
"step": 2005
},
{
"entropy": 6.1580390453338625,
"epoch": 0.156389807430461,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999304886650422,
"loss": 6.1388,
"mean_token_accuracy": 0.13328585550189018,
"num_tokens": 3499373.0,
"step": 2010
},
{
"entropy": 6.130110073089599,
"epoch": 0.15677883680217858,
"grad_norm": 1.125,
"learning_rate": 0.0004999297980809058,
"loss": 6.0894,
"mean_token_accuracy": 0.12947576269507408,
"num_tokens": 3508344.0,
"step": 2015
},
{
"entropy": 6.102607488632202,
"epoch": 0.15716786617389614,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999291040838176,
"loss": 6.0589,
"mean_token_accuracy": 0.12978120669722557,
"num_tokens": 3517009.0,
"step": 2020
},
{
"entropy": 6.054499006271362,
"epoch": 0.1575568955456137,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999284066737884,
"loss": 6.11,
"mean_token_accuracy": 0.12718966975808144,
"num_tokens": 3525989.0,
"step": 2025
},
{
"entropy": 6.133045625686646,
"epoch": 0.15794592491733125,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999277058508285,
"loss": 5.9964,
"mean_token_accuracy": 0.13161147758364677,
"num_tokens": 3533947.0,
"step": 2030
},
{
"entropy": 6.014222860336304,
"epoch": 0.1583349542890488,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999270016149486,
"loss": 6.0552,
"mean_token_accuracy": 0.135345246642828,
"num_tokens": 3541974.0,
"step": 2035
},
{
"entropy": 6.0979108810424805,
"epoch": 0.1587239836607664,
"grad_norm": 1.2734375,
"learning_rate": 0.0004999262939661596,
"loss": 6.0053,
"mean_token_accuracy": 0.14074506014585494,
"num_tokens": 3550436.0,
"step": 2040
},
{
"entropy": 6.0076916217803955,
"epoch": 0.15911301303248396,
"grad_norm": 1.140625,
"learning_rate": 0.0004999255829044721,
"loss": 5.9924,
"mean_token_accuracy": 0.13720230460166932,
"num_tokens": 3558283.0,
"step": 2045
},
{
"entropy": 5.98124508857727,
"epoch": 0.15950204240420152,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999248684298968,
"loss": 5.9144,
"mean_token_accuracy": 0.13598482608795165,
"num_tokens": 3566968.0,
"step": 2050
},
{
"entropy": 6.084092378616333,
"epoch": 0.15989107177591907,
"grad_norm": 1.109375,
"learning_rate": 0.0004999241505424447,
"loss": 6.0515,
"mean_token_accuracy": 0.13149691969156266,
"num_tokens": 3575580.0,
"step": 2055
},
{
"entropy": 6.089936590194702,
"epoch": 0.16028010114763663,
"grad_norm": 1.078125,
"learning_rate": 0.0004999234292421265,
"loss": 6.1059,
"mean_token_accuracy": 0.13127018511295319,
"num_tokens": 3584389.0,
"step": 2060
},
{
"entropy": 6.056532669067383,
"epoch": 0.16066913051935422,
"grad_norm": 1.09375,
"learning_rate": 0.0004999227045289535,
"loss": 5.9472,
"mean_token_accuracy": 0.13698288649320603,
"num_tokens": 3593009.0,
"step": 2065
},
{
"entropy": 5.953234529495239,
"epoch": 0.16105815989107178,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999219764029363,
"loss": 5.9771,
"mean_token_accuracy": 0.13605064973235131,
"num_tokens": 3601914.0,
"step": 2070
},
{
"entropy": 5.964067173004151,
"epoch": 0.16144718926278934,
"grad_norm": 1.15625,
"learning_rate": 0.0004999212448640861,
"loss": 5.9458,
"mean_token_accuracy": 0.1372133381664753,
"num_tokens": 3609883.0,
"step": 2075
},
{
"entropy": 5.984559869766235,
"epoch": 0.1618362186345069,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999205099124141,
"loss": 5.8776,
"mean_token_accuracy": 0.14128663763403893,
"num_tokens": 3618103.0,
"step": 2080
},
{
"entropy": 6.163353109359742,
"epoch": 0.16222524800622448,
"grad_norm": 1.140625,
"learning_rate": 0.0004999197715479313,
"loss": 6.1465,
"mean_token_accuracy": 0.13133149743080139,
"num_tokens": 3625861.0,
"step": 2085
},
{
"entropy": 5.983381605148315,
"epoch": 0.16261427737794204,
"grad_norm": 1.140625,
"learning_rate": 0.0004999190297706489,
"loss": 5.9377,
"mean_token_accuracy": 0.13726572021842004,
"num_tokens": 3634498.0,
"step": 2090
},
{
"entropy": 5.900870990753174,
"epoch": 0.1630033067496596,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999182845805783,
"loss": 5.903,
"mean_token_accuracy": 0.13446398749947547,
"num_tokens": 3644186.0,
"step": 2095
},
{
"entropy": 5.978477811813354,
"epoch": 0.16339233612137716,
"grad_norm": 1.15625,
"learning_rate": 0.0004999175359777306,
"loss": 6.0346,
"mean_token_accuracy": 0.13290069103240967,
"num_tokens": 3652831.0,
"step": 2100
},
{
"entropy": 6.1047296047210695,
"epoch": 0.16378136549309472,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999167839621174,
"loss": 5.9013,
"mean_token_accuracy": 0.13871957510709762,
"num_tokens": 3661230.0,
"step": 2105
},
{
"entropy": 6.000889015197754,
"epoch": 0.1641703948648123,
"grad_norm": 1.09375,
"learning_rate": 0.0004999160285337501,
"loss": 6.0459,
"mean_token_accuracy": 0.1324700802564621,
"num_tokens": 3669250.0,
"step": 2110
},
{
"entropy": 6.074883890151978,
"epoch": 0.16455942423652986,
"grad_norm": 1.140625,
"learning_rate": 0.0004999152696926399,
"loss": 6.0221,
"mean_token_accuracy": 0.1307499475777149,
"num_tokens": 3677568.0,
"step": 2115
},
{
"entropy": 6.026207399368286,
"epoch": 0.16494845360824742,
"grad_norm": 1.140625,
"learning_rate": 0.0004999145074387985,
"loss": 6.0186,
"mean_token_accuracy": 0.14040926843881607,
"num_tokens": 3685488.0,
"step": 2120
},
{
"entropy": 5.997472858428955,
"epoch": 0.16533748297996498,
"grad_norm": 1.125,
"learning_rate": 0.0004999137417722374,
"loss": 5.9989,
"mean_token_accuracy": 0.13259485363960266,
"num_tokens": 3694357.0,
"step": 2125
},
{
"entropy": 6.097675323486328,
"epoch": 0.16572651235168256,
"grad_norm": 1.1875,
"learning_rate": 0.0004999129726929684,
"loss": 6.0534,
"mean_token_accuracy": 0.1371625356376171,
"num_tokens": 3703173.0,
"step": 2130
},
{
"entropy": 6.020848321914673,
"epoch": 0.16611554172340012,
"grad_norm": 1.15625,
"learning_rate": 0.0004999122002010029,
"loss": 6.0185,
"mean_token_accuracy": 0.13592875599861146,
"num_tokens": 3712070.0,
"step": 2135
},
{
"entropy": 6.032235813140869,
"epoch": 0.16650457109511768,
"grad_norm": 1.21875,
"learning_rate": 0.0004999114242963527,
"loss": 5.9062,
"mean_token_accuracy": 0.14077977910637857,
"num_tokens": 3720245.0,
"step": 2140
},
{
"entropy": 5.971114015579223,
"epoch": 0.16689360046683524,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999106449790298,
"loss": 6.0256,
"mean_token_accuracy": 0.1356363706290722,
"num_tokens": 3730064.0,
"step": 2145
},
{
"entropy": 6.012196159362793,
"epoch": 0.1672826298385528,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999098622490458,
"loss": 5.9673,
"mean_token_accuracy": 0.13840274810791015,
"num_tokens": 3738741.0,
"step": 2150
},
{
"entropy": 6.0317319393157955,
"epoch": 0.16767165921027039,
"grad_norm": 1.125,
"learning_rate": 0.0004999090761064125,
"loss": 6.0445,
"mean_token_accuracy": 0.13798771649599076,
"num_tokens": 3747472.0,
"step": 2155
},
{
"entropy": 6.042453670501709,
"epoch": 0.16806068858198794,
"grad_norm": 1.140625,
"learning_rate": 0.000499908286551142,
"loss": 6.0614,
"mean_token_accuracy": 0.1325035110116005,
"num_tokens": 3756643.0,
"step": 2160
},
{
"entropy": 6.098313522338867,
"epoch": 0.1684497179537055,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999074935832463,
"loss": 6.1743,
"mean_token_accuracy": 0.1302264779806137,
"num_tokens": 3765490.0,
"step": 2165
},
{
"entropy": 6.109912395477295,
"epoch": 0.16883874732542306,
"grad_norm": 1.15625,
"learning_rate": 0.0004999066972027373,
"loss": 5.9293,
"mean_token_accuracy": 0.13885341733694076,
"num_tokens": 3773063.0,
"step": 2170
},
{
"entropy": 6.057087755203247,
"epoch": 0.16922777669714065,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999058974096271,
"loss": 6.081,
"mean_token_accuracy": 0.12848760411143303,
"num_tokens": 3781763.0,
"step": 2175
},
{
"entropy": 6.050684452056885,
"epoch": 0.1696168060688582,
"grad_norm": 1.125,
"learning_rate": 0.000499905094203928,
"loss": 6.0706,
"mean_token_accuracy": 0.13044458702206613,
"num_tokens": 3790582.0,
"step": 2180
},
{
"entropy": 5.997987556457519,
"epoch": 0.17000583544057576,
"grad_norm": 1.1484375,
"learning_rate": 0.000499904287585652,
"loss": 5.8916,
"mean_token_accuracy": 0.13762697726488113,
"num_tokens": 3798745.0,
"step": 2185
},
{
"entropy": 5.99886155128479,
"epoch": 0.17039486481229332,
"grad_norm": 1.109375,
"learning_rate": 0.0004999034775548113,
"loss": 5.9924,
"mean_token_accuracy": 0.13636661395430566,
"num_tokens": 3806901.0,
"step": 2190
},
{
"entropy": 6.0307700634002686,
"epoch": 0.17078389418401088,
"grad_norm": 1.328125,
"learning_rate": 0.0004999026641114185,
"loss": 5.9387,
"mean_token_accuracy": 0.14335163310170174,
"num_tokens": 3815742.0,
"step": 2195
},
{
"entropy": 5.951791095733642,
"epoch": 0.17117292355572847,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999018472554857,
"loss": 6.0196,
"mean_token_accuracy": 0.13228272050619125,
"num_tokens": 3824468.0,
"step": 2200
},
{
"entropy": 6.030127143859863,
"epoch": 0.17156195292744603,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999010269870253,
"loss": 5.9848,
"mean_token_accuracy": 0.14022162035107613,
"num_tokens": 3832834.0,
"step": 2205
},
{
"entropy": 6.018565368652344,
"epoch": 0.17195098229916359,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999002033060498,
"loss": 6.0239,
"mean_token_accuracy": 0.14021265134215355,
"num_tokens": 3841780.0,
"step": 2210
},
{
"entropy": 6.101462125778198,
"epoch": 0.17234001167088114,
"grad_norm": 1.125,
"learning_rate": 0.0004998993762125716,
"loss": 6.1403,
"mean_token_accuracy": 0.1279483050107956,
"num_tokens": 3849840.0,
"step": 2215
},
{
"entropy": 6.102045392990112,
"epoch": 0.17272904104259873,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998985457066034,
"loss": 6.0699,
"mean_token_accuracy": 0.12550090849399567,
"num_tokens": 3858204.0,
"step": 2220
},
{
"entropy": 6.012039756774902,
"epoch": 0.1731180704143163,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998977117881576,
"loss": 6.0052,
"mean_token_accuracy": 0.13568017780780792,
"num_tokens": 3866615.0,
"step": 2225
},
{
"entropy": 6.037418842315674,
"epoch": 0.17350709978603385,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998968744572472,
"loss": 6.0343,
"mean_token_accuracy": 0.13920788690447808,
"num_tokens": 3876003.0,
"step": 2230
},
{
"entropy": 6.0176434993743895,
"epoch": 0.1738961291577514,
"grad_norm": 1.140625,
"learning_rate": 0.0004998960337138845,
"loss": 5.9605,
"mean_token_accuracy": 0.1377207651734352,
"num_tokens": 3884362.0,
"step": 2235
},
{
"entropy": 5.990269565582276,
"epoch": 0.17428515852946896,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998951895580826,
"loss": 5.9158,
"mean_token_accuracy": 0.13989226818084716,
"num_tokens": 3893177.0,
"step": 2240
},
{
"entropy": 6.01600284576416,
"epoch": 0.17467418790118655,
"grad_norm": 1.2109375,
"learning_rate": 0.0004998943419898542,
"loss": 5.9773,
"mean_token_accuracy": 0.1380767658352852,
"num_tokens": 3903006.0,
"step": 2245
},
{
"entropy": 5.9797093868255615,
"epoch": 0.1750632172729041,
"grad_norm": 1.125,
"learning_rate": 0.0004998934910092119,
"loss": 5.8372,
"mean_token_accuracy": 0.13846131414175034,
"num_tokens": 3912256.0,
"step": 2250
},
{
"entropy": 5.916180229187011,
"epoch": 0.17545224664462167,
"grad_norm": 1.15625,
"learning_rate": 0.0004998926366161689,
"loss": 5.9986,
"mean_token_accuracy": 0.13596335276961327,
"num_tokens": 3921121.0,
"step": 2255
},
{
"entropy": 6.018570613861084,
"epoch": 0.17584127601633923,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998917788107383,
"loss": 5.9479,
"mean_token_accuracy": 0.139829271286726,
"num_tokens": 3929348.0,
"step": 2260
},
{
"entropy": 6.029901313781738,
"epoch": 0.17623030538805678,
"grad_norm": 1.140625,
"learning_rate": 0.0004998909175929326,
"loss": 6.1262,
"mean_token_accuracy": 0.13413718938827515,
"num_tokens": 3938073.0,
"step": 2265
},
{
"entropy": 6.116082048416137,
"epoch": 0.17661933475977437,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998900529627653,
"loss": 6.1048,
"mean_token_accuracy": 0.13597819358110427,
"num_tokens": 3946641.0,
"step": 2270
},
{
"entropy": 6.074714040756225,
"epoch": 0.17700836413149193,
"grad_norm": 1.109375,
"learning_rate": 0.0004998891849202495,
"loss": 5.9587,
"mean_token_accuracy": 0.14048855975270272,
"num_tokens": 3954586.0,
"step": 2275
},
{
"entropy": 6.009363508224487,
"epoch": 0.1773973935032095,
"grad_norm": 1.2109375,
"learning_rate": 0.000499888313465398,
"loss": 6.0295,
"mean_token_accuracy": 0.13712368234992028,
"num_tokens": 3963745.0,
"step": 2280
},
{
"entropy": 5.937222337722778,
"epoch": 0.17778642287492705,
"grad_norm": 1.171875,
"learning_rate": 0.0004998874385982247,
"loss": 5.8522,
"mean_token_accuracy": 0.1412455268204212,
"num_tokens": 3972452.0,
"step": 2285
},
{
"entropy": 5.94044041633606,
"epoch": 0.17817545224664463,
"grad_norm": 1.140625,
"learning_rate": 0.0004998865603187421,
"loss": 5.9574,
"mean_token_accuracy": 0.13704333379864692,
"num_tokens": 3980770.0,
"step": 2290
},
{
"entropy": 6.060723733901978,
"epoch": 0.1785644816183622,
"grad_norm": 1.0859375,
"learning_rate": 0.000499885678626964,
"loss": 5.9805,
"mean_token_accuracy": 0.13396030068397521,
"num_tokens": 3989545.0,
"step": 2295
},
{
"entropy": 6.062776803970337,
"epoch": 0.17895351099007975,
"grad_norm": 1.078125,
"learning_rate": 0.0004998847935229038,
"loss": 5.9424,
"mean_token_accuracy": 0.13457537293434144,
"num_tokens": 3998034.0,
"step": 2300
},
{
"entropy": 5.905241298675537,
"epoch": 0.1793425403617973,
"grad_norm": 1.140625,
"learning_rate": 0.0004998839050065746,
"loss": 5.9822,
"mean_token_accuracy": 0.14144275784492494,
"num_tokens": 4007110.0,
"step": 2305
},
{
"entropy": 6.042173862457275,
"epoch": 0.17973156973351487,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998830130779902,
"loss": 6.0362,
"mean_token_accuracy": 0.12785631343722342,
"num_tokens": 4015930.0,
"step": 2310
},
{
"entropy": 5.92177267074585,
"epoch": 0.18012059910523245,
"grad_norm": 1.109375,
"learning_rate": 0.000499882117737164,
"loss": 5.8567,
"mean_token_accuracy": 0.13938324376940728,
"num_tokens": 4024510.0,
"step": 2315
},
{
"entropy": 5.9428199291229244,
"epoch": 0.18050962847695,
"grad_norm": 1.09375,
"learning_rate": 0.0004998812189841096,
"loss": 5.9629,
"mean_token_accuracy": 0.13904105722904206,
"num_tokens": 4033714.0,
"step": 2320
},
{
"entropy": 6.135042381286621,
"epoch": 0.18089865784866757,
"grad_norm": 1.1875,
"learning_rate": 0.0004998803168188407,
"loss": 6.1461,
"mean_token_accuracy": 0.12715203911066056,
"num_tokens": 4042376.0,
"step": 2325
},
{
"entropy": 6.03497052192688,
"epoch": 0.18128768722038513,
"grad_norm": 1.125,
"learning_rate": 0.0004998794112413708,
"loss": 5.9316,
"mean_token_accuracy": 0.1389988273382187,
"num_tokens": 4050521.0,
"step": 2330
},
{
"entropy": 6.097191667556762,
"epoch": 0.18167671659210272,
"grad_norm": 1.109375,
"learning_rate": 0.0004998785022517137,
"loss": 6.0873,
"mean_token_accuracy": 0.13467619195580482,
"num_tokens": 4059384.0,
"step": 2335
},
{
"entropy": 5.945719146728516,
"epoch": 0.18206574596382027,
"grad_norm": 1.203125,
"learning_rate": 0.0004998775898498835,
"loss": 5.947,
"mean_token_accuracy": 0.1388249836862087,
"num_tokens": 4067655.0,
"step": 2340
},
{
"entropy": 6.028331184387207,
"epoch": 0.18245477533553783,
"grad_norm": 1.140625,
"learning_rate": 0.0004998766740358936,
"loss": 5.9342,
"mean_token_accuracy": 0.1423063322901726,
"num_tokens": 4075828.0,
"step": 2345
},
{
"entropy": 6.014834260940551,
"epoch": 0.1828438047072554,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998757548097582,
"loss": 5.9053,
"mean_token_accuracy": 0.14048654288053514,
"num_tokens": 4083796.0,
"step": 2350
},
{
"entropy": 5.980800151824951,
"epoch": 0.18323283407897295,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998748321714911,
"loss": 6.0421,
"mean_token_accuracy": 0.13735400661826133,
"num_tokens": 4092981.0,
"step": 2355
},
{
"entropy": 6.020400857925415,
"epoch": 0.18362186345069054,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998739061211065,
"loss": 5.8484,
"mean_token_accuracy": 0.14194852709770203,
"num_tokens": 4101363.0,
"step": 2360
},
{
"entropy": 5.96309461593628,
"epoch": 0.1840108928224081,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998729766586181,
"loss": 5.9272,
"mean_token_accuracy": 0.1333435907959938,
"num_tokens": 4110164.0,
"step": 2365
},
{
"entropy": 6.044630527496338,
"epoch": 0.18439992219412565,
"grad_norm": 1.15625,
"learning_rate": 0.0004998720437840403,
"loss": 6.0444,
"mean_token_accuracy": 0.13811287730932237,
"num_tokens": 4119098.0,
"step": 2370
},
{
"entropy": 5.991950607299804,
"epoch": 0.1847889515658432,
"grad_norm": 1.21875,
"learning_rate": 0.0004998711074973871,
"loss": 5.9532,
"mean_token_accuracy": 0.13917487487196922,
"num_tokens": 4127522.0,
"step": 2375
},
{
"entropy": 6.0055859088897705,
"epoch": 0.1851779809375608,
"grad_norm": 1.125,
"learning_rate": 0.0004998701677986728,
"loss": 5.9803,
"mean_token_accuracy": 0.14277126044034957,
"num_tokens": 4136311.0,
"step": 2380
},
{
"entropy": 5.935305976867676,
"epoch": 0.18556701030927836,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998692246879116,
"loss": 5.9124,
"mean_token_accuracy": 0.13691533505916595,
"num_tokens": 4145160.0,
"step": 2385
},
{
"entropy": 6.005577945709229,
"epoch": 0.18595603968099592,
"grad_norm": 1.1875,
"learning_rate": 0.0004998682781651178,
"loss": 5.886,
"mean_token_accuracy": 0.14444762617349624,
"num_tokens": 4153921.0,
"step": 2390
},
{
"entropy": 5.912683391571045,
"epoch": 0.18634506905271347,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998673282303059,
"loss": 5.9582,
"mean_token_accuracy": 0.13400443345308305,
"num_tokens": 4162580.0,
"step": 2395
},
{
"entropy": 5.9940591812133786,
"epoch": 0.18673409842443103,
"grad_norm": 1.109375,
"learning_rate": 0.0004998663748834901,
"loss": 5.9552,
"mean_token_accuracy": 0.14112788587808608,
"num_tokens": 4170623.0,
"step": 2400
},
{
"entropy": 5.947180700302124,
"epoch": 0.18712312779614862,
"grad_norm": 1.140625,
"learning_rate": 0.0004998654181246851,
"loss": 5.9519,
"mean_token_accuracy": 0.14752146303653718,
"num_tokens": 4178724.0,
"step": 2405
},
{
"entropy": 5.964427280426025,
"epoch": 0.18751215716786618,
"grad_norm": 1.140625,
"learning_rate": 0.0004998644579539052,
"loss": 5.963,
"mean_token_accuracy": 0.13656028136610984,
"num_tokens": 4187773.0,
"step": 2410
},
{
"entropy": 6.009492588043213,
"epoch": 0.18790118653958374,
"grad_norm": 1.109375,
"learning_rate": 0.000499863494371165,
"loss": 5.9618,
"mean_token_accuracy": 0.13550449684262275,
"num_tokens": 4196628.0,
"step": 2415
},
{
"entropy": 5.984348201751709,
"epoch": 0.1882902159113013,
"grad_norm": 1.203125,
"learning_rate": 0.0004998625273764793,
"loss": 5.9827,
"mean_token_accuracy": 0.13410525098443032,
"num_tokens": 4205090.0,
"step": 2420
},
{
"entropy": 5.990939712524414,
"epoch": 0.18867924528301888,
"grad_norm": 1.15625,
"learning_rate": 0.0004998615569698626,
"loss": 5.8965,
"mean_token_accuracy": 0.1415567897260189,
"num_tokens": 4212757.0,
"step": 2425
},
{
"entropy": 5.883493232727051,
"epoch": 0.18906827465473644,
"grad_norm": 1.203125,
"learning_rate": 0.0004998605831513296,
"loss": 5.9637,
"mean_token_accuracy": 0.14001091420650483,
"num_tokens": 4221556.0,
"step": 2430
},
{
"entropy": 5.936648797988892,
"epoch": 0.189457304026454,
"grad_norm": 1.203125,
"learning_rate": 0.0004998596059208953,
"loss": 5.9587,
"mean_token_accuracy": 0.13833684399724006,
"num_tokens": 4231524.0,
"step": 2435
},
{
"entropy": 6.036439371109009,
"epoch": 0.18984633339817156,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998586252785743,
"loss": 5.9826,
"mean_token_accuracy": 0.1431053563952446,
"num_tokens": 4240229.0,
"step": 2440
},
{
"entropy": 5.956485605239868,
"epoch": 0.19023536276988912,
"grad_norm": 1.1875,
"learning_rate": 0.0004998576412243816,
"loss": 5.8587,
"mean_token_accuracy": 0.14437367022037506,
"num_tokens": 4248053.0,
"step": 2445
},
{
"entropy": 5.988489770889283,
"epoch": 0.1906243921416067,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998566537583321,
"loss": 5.9007,
"mean_token_accuracy": 0.13747418969869613,
"num_tokens": 4256019.0,
"step": 2450
},
{
"entropy": 5.89534101486206,
"epoch": 0.19101342151332426,
"grad_norm": 1.203125,
"learning_rate": 0.0004998556628804408,
"loss": 5.8917,
"mean_token_accuracy": 0.14226853474974632,
"num_tokens": 4263805.0,
"step": 2455
},
{
"entropy": 5.958291482925415,
"epoch": 0.19140245088504182,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998546685907225,
"loss": 5.9534,
"mean_token_accuracy": 0.13563379049301147,
"num_tokens": 4272712.0,
"step": 2460
},
{
"entropy": 6.066695690155029,
"epoch": 0.19179148025675938,
"grad_norm": 1.140625,
"learning_rate": 0.0004998536708891927,
"loss": 5.8824,
"mean_token_accuracy": 0.14191587790846824,
"num_tokens": 4281222.0,
"step": 2465
},
{
"entropy": 5.9650391101837155,
"epoch": 0.19218050962847694,
"grad_norm": 1.15625,
"learning_rate": 0.0004998526697758663,
"loss": 6.0121,
"mean_token_accuracy": 0.13257578536868095,
"num_tokens": 4290558.0,
"step": 2470
},
{
"entropy": 6.109343481063843,
"epoch": 0.19256953900019452,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998516652507585,
"loss": 6.0253,
"mean_token_accuracy": 0.13303112387657165,
"num_tokens": 4299684.0,
"step": 2475
},
{
"entropy": 5.899087524414062,
"epoch": 0.19295856837191208,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998506573138846,
"loss": 5.9633,
"mean_token_accuracy": 0.14104458689689636,
"num_tokens": 4308938.0,
"step": 2480
},
{
"entropy": 5.9440470218658445,
"epoch": 0.19334759774362964,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998496459652598,
"loss": 5.8208,
"mean_token_accuracy": 0.14647359624505044,
"num_tokens": 4317646.0,
"step": 2485
},
{
"entropy": 5.964761638641358,
"epoch": 0.1937366271153472,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998486312048995,
"loss": 5.9621,
"mean_token_accuracy": 0.13583191409707068,
"num_tokens": 4326876.0,
"step": 2490
},
{
"entropy": 5.967274475097656,
"epoch": 0.19412565648706478,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998476130328193,
"loss": 5.931,
"mean_token_accuracy": 0.13568915724754332,
"num_tokens": 4335391.0,
"step": 2495
},
{
"entropy": 6.055395555496216,
"epoch": 0.19451468585878234,
"grad_norm": 1.140625,
"learning_rate": 0.0004998465914490343,
"loss": 5.9052,
"mean_token_accuracy": 0.1406179554760456,
"num_tokens": 4343253.0,
"step": 2500
},
{
"entropy": 5.760255670547485,
"epoch": 0.1949037152304999,
"grad_norm": 1.0546875,
"learning_rate": 0.00049984556645356,
"loss": 5.8742,
"mean_token_accuracy": 0.14200342074036598,
"num_tokens": 4352161.0,
"step": 2505
},
{
"entropy": 5.938892364501953,
"epoch": 0.19529274460221746,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998445380464124,
"loss": 5.8385,
"mean_token_accuracy": 0.1466829337179661,
"num_tokens": 4360414.0,
"step": 2510
},
{
"entropy": 5.896971416473389,
"epoch": 0.19568177397393502,
"grad_norm": 1.15625,
"learning_rate": 0.0004998435062276066,
"loss": 5.8228,
"mean_token_accuracy": 0.14513540267944336,
"num_tokens": 4369375.0,
"step": 2515
},
{
"entropy": 5.903877830505371,
"epoch": 0.1960708033456526,
"grad_norm": 1.1953125,
"learning_rate": 0.0004998424709971586,
"loss": 5.8483,
"mean_token_accuracy": 0.14223146587610244,
"num_tokens": 4378638.0,
"step": 2520
},
{
"entropy": 5.983705711364746,
"epoch": 0.19645983271737016,
"grad_norm": 1.171875,
"learning_rate": 0.0004998414323550839,
"loss": 5.8616,
"mean_token_accuracy": 0.14103932306170464,
"num_tokens": 4388262.0,
"step": 2525
},
{
"entropy": 5.874213314056396,
"epoch": 0.19684886208908772,
"grad_norm": 1.109375,
"learning_rate": 0.0004998403903013984,
"loss": 5.8501,
"mean_token_accuracy": 0.1444000080227852,
"num_tokens": 4396894.0,
"step": 2530
},
{
"entropy": 5.949297761917114,
"epoch": 0.19723789146080528,
"grad_norm": 1.03125,
"learning_rate": 0.0004998393448361179,
"loss": 6.0004,
"mean_token_accuracy": 0.1377578116953373,
"num_tokens": 4405455.0,
"step": 2535
},
{
"entropy": 5.937081432342529,
"epoch": 0.19762692083252287,
"grad_norm": 1.359375,
"learning_rate": 0.000499838295959258,
"loss": 5.8999,
"mean_token_accuracy": 0.13834612667560578,
"num_tokens": 4414355.0,
"step": 2540
},
{
"entropy": 5.8034745216369625,
"epoch": 0.19801595020424043,
"grad_norm": 1.1875,
"learning_rate": 0.000499837243670835,
"loss": 5.7847,
"mean_token_accuracy": 0.14933302402496337,
"num_tokens": 4422305.0,
"step": 2545
},
{
"entropy": 5.9695093631744385,
"epoch": 0.19840497957595798,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998361879708646,
"loss": 5.9185,
"mean_token_accuracy": 0.13912818506360053,
"num_tokens": 4430813.0,
"step": 2550
},
{
"entropy": 5.963333559036255,
"epoch": 0.19879400894767554,
"grad_norm": 1.09375,
"learning_rate": 0.000499835128859363,
"loss": 5.8774,
"mean_token_accuracy": 0.14088211506605147,
"num_tokens": 4440229.0,
"step": 2555
},
{
"entropy": 5.8513782024383545,
"epoch": 0.1991830383193931,
"grad_norm": 1.140625,
"learning_rate": 0.0004998340663363461,
"loss": 5.9025,
"mean_token_accuracy": 0.13724920377135277,
"num_tokens": 4449417.0,
"step": 2560
},
{
"entropy": 5.9853545188903805,
"epoch": 0.1995720676911107,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998330004018301,
"loss": 5.944,
"mean_token_accuracy": 0.13790054023265838,
"num_tokens": 4457377.0,
"step": 2565
},
{
"entropy": 5.990766906738282,
"epoch": 0.19996109706282825,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998319310558312,
"loss": 5.8617,
"mean_token_accuracy": 0.14069739058613778,
"num_tokens": 4466839.0,
"step": 2570
},
{
"entropy": 5.904648494720459,
"epoch": 0.2003501264345458,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998308582983657,
"loss": 5.9051,
"mean_token_accuracy": 0.14068260937929153,
"num_tokens": 4476369.0,
"step": 2575
},
{
"entropy": 5.9384589195251465,
"epoch": 0.20073915580626336,
"grad_norm": 1.125,
"learning_rate": 0.0004998297821294497,
"loss": 5.9575,
"mean_token_accuracy": 0.13891851380467415,
"num_tokens": 4485609.0,
"step": 2580
},
{
"entropy": 5.939151048660278,
"epoch": 0.20112818517798095,
"grad_norm": 1.171875,
"learning_rate": 0.0004998287025490995,
"loss": 5.9092,
"mean_token_accuracy": 0.1407192036509514,
"num_tokens": 4494535.0,
"step": 2585
},
{
"entropy": 5.945156764984131,
"epoch": 0.2015172145496985,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998276195573317,
"loss": 5.8844,
"mean_token_accuracy": 0.1363791175186634,
"num_tokens": 4503133.0,
"step": 2590
},
{
"entropy": 5.885627603530883,
"epoch": 0.20190624392141607,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998265331541627,
"loss": 5.7876,
"mean_token_accuracy": 0.15038875490427017,
"num_tokens": 4511465.0,
"step": 2595
},
{
"entropy": 5.883782386779785,
"epoch": 0.20229527329313363,
"grad_norm": 1.2109375,
"learning_rate": 0.0004998254433396088,
"loss": 5.8499,
"mean_token_accuracy": 0.14612346962094308,
"num_tokens": 4520177.0,
"step": 2600
},
{
"entropy": 5.947019147872925,
"epoch": 0.20268430266485118,
"grad_norm": 1.1875,
"learning_rate": 0.0004998243501136867,
"loss": 6.0221,
"mean_token_accuracy": 0.13749735280871392,
"num_tokens": 4528370.0,
"step": 2605
},
{
"entropy": 6.0327956199646,
"epoch": 0.20307333203656877,
"grad_norm": 1.2109375,
"learning_rate": 0.0004998232534764129,
"loss": 6.0424,
"mean_token_accuracy": 0.13159574791789055,
"num_tokens": 4538258.0,
"step": 2610
},
{
"entropy": 5.933379316329956,
"epoch": 0.20346236140828633,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998221534278041,
"loss": 5.9334,
"mean_token_accuracy": 0.13688500002026557,
"num_tokens": 4547300.0,
"step": 2615
},
{
"entropy": 6.0140650272369385,
"epoch": 0.2038513907800039,
"grad_norm": 1.015625,
"learning_rate": 0.0004998210499678769,
"loss": 5.9854,
"mean_token_accuracy": 0.1308668576180935,
"num_tokens": 4556877.0,
"step": 2620
},
{
"entropy": 5.907776832580566,
"epoch": 0.20424042015172145,
"grad_norm": 1.125,
"learning_rate": 0.0004998199430966481,
"loss": 5.8302,
"mean_token_accuracy": 0.14112840071320534,
"num_tokens": 4565999.0,
"step": 2625
},
{
"entropy": 5.913699102401734,
"epoch": 0.20462944952343903,
"grad_norm": 1.1875,
"learning_rate": 0.0004998188328141346,
"loss": 5.8793,
"mean_token_accuracy": 0.13416805639863014,
"num_tokens": 4574027.0,
"step": 2630
},
{
"entropy": 5.887786960601806,
"epoch": 0.2050184788951566,
"grad_norm": 1.15625,
"learning_rate": 0.0004998177191203531,
"loss": 5.8994,
"mean_token_accuracy": 0.1329669289290905,
"num_tokens": 4582402.0,
"step": 2635
},
{
"entropy": 5.92496690750122,
"epoch": 0.20540750826687415,
"grad_norm": 1.2734375,
"learning_rate": 0.0004998166020153204,
"loss": 5.8358,
"mean_token_accuracy": 0.14105968326330184,
"num_tokens": 4590381.0,
"step": 2640
},
{
"entropy": 5.917643356323242,
"epoch": 0.2057965376385917,
"grad_norm": 1.2578125,
"learning_rate": 0.0004998154814990538,
"loss": 5.885,
"mean_token_accuracy": 0.14099830090999604,
"num_tokens": 4598910.0,
"step": 2645
},
{
"entropy": 6.011348581314087,
"epoch": 0.20618556701030927,
"grad_norm": 1.1953125,
"learning_rate": 0.00049981435757157,
"loss": 6.015,
"mean_token_accuracy": 0.13125852718949318,
"num_tokens": 4607400.0,
"step": 2650
},
{
"entropy": 5.917462635040283,
"epoch": 0.20657459638202685,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998132302328862,
"loss": 5.9403,
"mean_token_accuracy": 0.14050790518522263,
"num_tokens": 4616510.0,
"step": 2655
},
{
"entropy": 6.007234954833985,
"epoch": 0.2069636257537444,
"grad_norm": 1.2109375,
"learning_rate": 0.0004998120994830195,
"loss": 5.9072,
"mean_token_accuracy": 0.14376826882362365,
"num_tokens": 4624529.0,
"step": 2660
},
{
"entropy": 5.918785476684571,
"epoch": 0.20735265512546197,
"grad_norm": 1.1640625,
"learning_rate": 0.000499810965321987,
"loss": 5.8346,
"mean_token_accuracy": 0.14414578527212143,
"num_tokens": 4632141.0,
"step": 2665
},
{
"entropy": 5.9577617168426515,
"epoch": 0.20774168449717953,
"grad_norm": 1.21875,
"learning_rate": 0.0004998098277498059,
"loss": 6.0381,
"mean_token_accuracy": 0.1350015841424465,
"num_tokens": 4640378.0,
"step": 2670
},
{
"entropy": 5.925170755386352,
"epoch": 0.2081307138688971,
"grad_norm": 1.09375,
"learning_rate": 0.0004998086867664934,
"loss": 5.8409,
"mean_token_accuracy": 0.1414250299334526,
"num_tokens": 4649313.0,
"step": 2675
},
{
"entropy": 5.9586829662323,
"epoch": 0.20851974324061467,
"grad_norm": 1.203125,
"learning_rate": 0.000499807542372067,
"loss": 5.9164,
"mean_token_accuracy": 0.14138886630535125,
"num_tokens": 4658272.0,
"step": 2680
},
{
"entropy": 5.856334686279297,
"epoch": 0.20890877261233223,
"grad_norm": 1.203125,
"learning_rate": 0.0004998063945665439,
"loss": 5.7767,
"mean_token_accuracy": 0.14743116796016692,
"num_tokens": 4665970.0,
"step": 2685
},
{
"entropy": 5.974782228469849,
"epoch": 0.2092978019840498,
"grad_norm": 1.125,
"learning_rate": 0.0004998052433499416,
"loss": 6.0326,
"mean_token_accuracy": 0.13459649160504342,
"num_tokens": 4675308.0,
"step": 2690
},
{
"entropy": 5.90858154296875,
"epoch": 0.20968683135576735,
"grad_norm": 1.203125,
"learning_rate": 0.0004998040887222776,
"loss": 5.8772,
"mean_token_accuracy": 0.14054309129714965,
"num_tokens": 4684316.0,
"step": 2695
},
{
"entropy": 5.852538108825684,
"epoch": 0.21007586072748494,
"grad_norm": 1.234375,
"learning_rate": 0.0004998029306835693,
"loss": 5.7932,
"mean_token_accuracy": 0.1447305828332901,
"num_tokens": 4693029.0,
"step": 2700
},
{
"entropy": 5.91006441116333,
"epoch": 0.2104648900992025,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998017692338344,
"loss": 5.9919,
"mean_token_accuracy": 0.1333250030875206,
"num_tokens": 4702404.0,
"step": 2705
},
{
"entropy": 5.972537851333618,
"epoch": 0.21085391947092005,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998006043730905,
"loss": 5.9484,
"mean_token_accuracy": 0.13651543334126473,
"num_tokens": 4711126.0,
"step": 2710
},
{
"entropy": 5.994960069656372,
"epoch": 0.2112429488426376,
"grad_norm": 1.09375,
"learning_rate": 0.0004997994361013551,
"loss": 5.9405,
"mean_token_accuracy": 0.14613958448171616,
"num_tokens": 4719816.0,
"step": 2715
},
{
"entropy": 5.966370105743408,
"epoch": 0.21163197821435517,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997982644186461,
"loss": 5.8798,
"mean_token_accuracy": 0.13973141759634017,
"num_tokens": 4728294.0,
"step": 2720
},
{
"entropy": 5.888756990432739,
"epoch": 0.21202100758607276,
"grad_norm": 1.109375,
"learning_rate": 0.0004997970893249813,
"loss": 5.8275,
"mean_token_accuracy": 0.1435457229614258,
"num_tokens": 4736708.0,
"step": 2725
},
{
"entropy": 5.873240041732788,
"epoch": 0.21241003695779032,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997959108203785,
"loss": 6.0039,
"mean_token_accuracy": 0.13070807754993438,
"num_tokens": 4744893.0,
"step": 2730
},
{
"entropy": 6.096560430526734,
"epoch": 0.21279906632950787,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997947289048554,
"loss": 5.9623,
"mean_token_accuracy": 0.13920049890875816,
"num_tokens": 4753246.0,
"step": 2735
},
{
"entropy": 5.931299209594727,
"epoch": 0.21318809570122543,
"grad_norm": 1.140625,
"learning_rate": 0.0004997935435784302,
"loss": 5.8833,
"mean_token_accuracy": 0.13352058157324792,
"num_tokens": 4762229.0,
"step": 2740
},
{
"entropy": 5.805833244323731,
"epoch": 0.21357712507294302,
"grad_norm": 1.1328125,
"learning_rate": 0.0004997923548411208,
"loss": 5.8366,
"mean_token_accuracy": 0.14200862497091293,
"num_tokens": 4771315.0,
"step": 2745
},
{
"entropy": 5.755913734436035,
"epoch": 0.21396615444466058,
"grad_norm": 1.1796875,
"learning_rate": 0.000499791162692945,
"loss": 5.7604,
"mean_token_accuracy": 0.15337736159563065,
"num_tokens": 4779848.0,
"step": 2750
},
{
"entropy": 5.984225654602051,
"epoch": 0.21435518381637814,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997899671339214,
"loss": 6.0054,
"mean_token_accuracy": 0.14065569192171096,
"num_tokens": 4788939.0,
"step": 2755
},
{
"entropy": 5.942894124984742,
"epoch": 0.2147442131880957,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997887681640675,
"loss": 5.906,
"mean_token_accuracy": 0.13755589798092843,
"num_tokens": 4797719.0,
"step": 2760
},
{
"entropy": 5.928773069381714,
"epoch": 0.21513324255981325,
"grad_norm": 1.2109375,
"learning_rate": 0.0004997875657834021,
"loss": 5.9602,
"mean_token_accuracy": 0.1399763911962509,
"num_tokens": 4805601.0,
"step": 2765
},
{
"entropy": 6.0012726306915285,
"epoch": 0.21552227193153084,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997863599919432,
"loss": 5.9298,
"mean_token_accuracy": 0.13780287876725197,
"num_tokens": 4814993.0,
"step": 2770
},
{
"entropy": 5.983278465270996,
"epoch": 0.2159113013032484,
"grad_norm": 1.171875,
"learning_rate": 0.0004997851507897089,
"loss": 5.8699,
"mean_token_accuracy": 0.14026867374777793,
"num_tokens": 4824156.0,
"step": 2775
},
{
"entropy": 5.871163606643677,
"epoch": 0.21630033067496596,
"grad_norm": 1.15625,
"learning_rate": 0.0004997839381767178,
"loss": 5.9559,
"mean_token_accuracy": 0.13493574261665345,
"num_tokens": 4833827.0,
"step": 2780
},
{
"entropy": 5.923353576660157,
"epoch": 0.21668936004668352,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997827221529882,
"loss": 5.9086,
"mean_token_accuracy": 0.14079293385148048,
"num_tokens": 4842801.0,
"step": 2785
},
{
"entropy": 5.858642721176148,
"epoch": 0.2170783894184011,
"grad_norm": 1.1875,
"learning_rate": 0.0004997815027185386,
"loss": 5.7611,
"mean_token_accuracy": 0.14427610188722612,
"num_tokens": 4850928.0,
"step": 2790
},
{
"entropy": 5.874434089660644,
"epoch": 0.21746741879011866,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997802798733874,
"loss": 5.8683,
"mean_token_accuracy": 0.14393499940633775,
"num_tokens": 4859510.0,
"step": 2795
},
{
"entropy": 5.975643491744995,
"epoch": 0.21785644816183622,
"grad_norm": 1.015625,
"learning_rate": 0.0004997790536175534,
"loss": 5.9173,
"mean_token_accuracy": 0.14145495295524596,
"num_tokens": 4868201.0,
"step": 2800
},
{
"entropy": 5.86905689239502,
"epoch": 0.21824547753355378,
"grad_norm": 1.140625,
"learning_rate": 0.0004997778239510548,
"loss": 5.8335,
"mean_token_accuracy": 0.14240052700042724,
"num_tokens": 4876896.0,
"step": 2805
},
{
"entropy": 5.826552057266236,
"epoch": 0.21863450690527134,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997765908739106,
"loss": 5.7898,
"mean_token_accuracy": 0.15225823372602462,
"num_tokens": 4885136.0,
"step": 2810
},
{
"entropy": 5.826451778411865,
"epoch": 0.21902353627698892,
"grad_norm": 1.203125,
"learning_rate": 0.0004997753543861395,
"loss": 5.8053,
"mean_token_accuracy": 0.1449057564139366,
"num_tokens": 4893277.0,
"step": 2815
},
{
"entropy": 5.87442626953125,
"epoch": 0.21941256564870648,
"grad_norm": 1.1328125,
"learning_rate": 0.00049977411448776,
"loss": 5.962,
"mean_token_accuracy": 0.1370394378900528,
"num_tokens": 4901977.0,
"step": 2820
},
{
"entropy": 6.066293048858642,
"epoch": 0.21980159502042404,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997728711787912,
"loss": 5.9498,
"mean_token_accuracy": 0.13320593908429146,
"num_tokens": 4910888.0,
"step": 2825
},
{
"entropy": 5.899124479293823,
"epoch": 0.2201906243921416,
"grad_norm": 1.125,
"learning_rate": 0.0004997716244592518,
"loss": 5.9556,
"mean_token_accuracy": 0.13331344723701477,
"num_tokens": 4920369.0,
"step": 2830
},
{
"entropy": 5.970640707015991,
"epoch": 0.22057965376385918,
"grad_norm": 1.2578125,
"learning_rate": 0.0004997703743291607,
"loss": 5.9372,
"mean_token_accuracy": 0.13818638548254966,
"num_tokens": 4928799.0,
"step": 2835
},
{
"entropy": 5.959006452560425,
"epoch": 0.22096868313557674,
"grad_norm": 1.140625,
"learning_rate": 0.000499769120788537,
"loss": 5.8186,
"mean_token_accuracy": 0.14796194583177566,
"num_tokens": 4937194.0,
"step": 2840
},
{
"entropy": 5.892886638641357,
"epoch": 0.2213577125072943,
"grad_norm": 1.2265625,
"learning_rate": 0.0004997678638373995,
"loss": 5.9246,
"mean_token_accuracy": 0.13645815178751947,
"num_tokens": 4946174.0,
"step": 2845
},
{
"entropy": 5.890604877471924,
"epoch": 0.22174674187901186,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997666034757676,
"loss": 5.8636,
"mean_token_accuracy": 0.1371280185878277,
"num_tokens": 4955376.0,
"step": 2850
},
{
"entropy": 5.896667575836181,
"epoch": 0.22213577125072942,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997653397036603,
"loss": 5.8672,
"mean_token_accuracy": 0.14396676495671273,
"num_tokens": 4964138.0,
"step": 2855
},
{
"entropy": 5.889484786987305,
"epoch": 0.222524800622447,
"grad_norm": 1.109375,
"learning_rate": 0.0004997640725210965,
"loss": 5.9614,
"mean_token_accuracy": 0.13742173463106155,
"num_tokens": 4973868.0,
"step": 2860
},
{
"entropy": 5.917964696884155,
"epoch": 0.22291382999416456,
"grad_norm": 1.28125,
"learning_rate": 0.0004997628019280958,
"loss": 5.8382,
"mean_token_accuracy": 0.1428116887807846,
"num_tokens": 4982687.0,
"step": 2865
},
{
"entropy": 5.931079006195068,
"epoch": 0.22330285936588212,
"grad_norm": 1.1953125,
"learning_rate": 0.0004997615279246773,
"loss": 5.8865,
"mean_token_accuracy": 0.13743261098861695,
"num_tokens": 4992072.0,
"step": 2870
},
{
"entropy": 5.911539316177368,
"epoch": 0.22369188873759968,
"grad_norm": 1.125,
"learning_rate": 0.0004997602505108603,
"loss": 5.8446,
"mean_token_accuracy": 0.1401855692267418,
"num_tokens": 5000551.0,
"step": 2875
},
{
"entropy": 5.937168121337891,
"epoch": 0.22408091810931724,
"grad_norm": 1.140625,
"learning_rate": 0.0004997589696866644,
"loss": 5.9994,
"mean_token_accuracy": 0.13774950057268143,
"num_tokens": 5009292.0,
"step": 2880
},
{
"entropy": 5.91554889678955,
"epoch": 0.22446994748103483,
"grad_norm": 1.125,
"learning_rate": 0.0004997576854521088,
"loss": 5.8535,
"mean_token_accuracy": 0.13967245817184448,
"num_tokens": 5018407.0,
"step": 2885
},
{
"entropy": 5.8885914325714115,
"epoch": 0.22485897685275238,
"grad_norm": 1.1171875,
"learning_rate": 0.000499756397807213,
"loss": 5.9008,
"mean_token_accuracy": 0.1394205868244171,
"num_tokens": 5027353.0,
"step": 2890
},
{
"entropy": 6.03417592048645,
"epoch": 0.22524800622446994,
"grad_norm": 1.140625,
"learning_rate": 0.0004997551067519966,
"loss": 5.9215,
"mean_token_accuracy": 0.13638510555028915,
"num_tokens": 5036476.0,
"step": 2895
},
{
"entropy": 5.861401844024658,
"epoch": 0.2256370355961875,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997538122864792,
"loss": 5.8525,
"mean_token_accuracy": 0.13833182156085969,
"num_tokens": 5045132.0,
"step": 2900
},
{
"entropy": 5.916049289703369,
"epoch": 0.2260260649679051,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997525144106804,
"loss": 5.872,
"mean_token_accuracy": 0.14037140905857087,
"num_tokens": 5053665.0,
"step": 2905
},
{
"entropy": 5.922544813156128,
"epoch": 0.22641509433962265,
"grad_norm": 1.109375,
"learning_rate": 0.00049975121312462,
"loss": 5.8271,
"mean_token_accuracy": 0.14556183218955993,
"num_tokens": 5062516.0,
"step": 2910
},
{
"entropy": 5.899590253829956,
"epoch": 0.2268041237113402,
"grad_norm": 1.203125,
"learning_rate": 0.0004997499084283177,
"loss": 5.8772,
"mean_token_accuracy": 0.13606662452220916,
"num_tokens": 5071260.0,
"step": 2915
},
{
"entropy": 5.827817678451538,
"epoch": 0.22719315308305776,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997486003217932,
"loss": 5.8196,
"mean_token_accuracy": 0.1497701920568943,
"num_tokens": 5079215.0,
"step": 2920
},
{
"entropy": 5.8816290378570555,
"epoch": 0.22758218245477532,
"grad_norm": 1.140625,
"learning_rate": 0.0004997472888050664,
"loss": 5.8938,
"mean_token_accuracy": 0.14027459546923637,
"num_tokens": 5088404.0,
"step": 2925
},
{
"entropy": 5.900744438171387,
"epoch": 0.2279712118264929,
"grad_norm": 1.2109375,
"learning_rate": 0.0004997459738781573,
"loss": 5.8239,
"mean_token_accuracy": 0.1431146442890167,
"num_tokens": 5096756.0,
"step": 2930
},
{
"entropy": 5.979656934738159,
"epoch": 0.22836024119821047,
"grad_norm": 1.078125,
"learning_rate": 0.0004997446555410857,
"loss": 5.9904,
"mean_token_accuracy": 0.13891934603452682,
"num_tokens": 5106160.0,
"step": 2935
},
{
"entropy": 5.7854640007019045,
"epoch": 0.22874927056992803,
"grad_norm": 1.15625,
"learning_rate": 0.0004997433337938716,
"loss": 5.9235,
"mean_token_accuracy": 0.14122586995363234,
"num_tokens": 5114915.0,
"step": 2940
},
{
"entropy": 5.964998531341553,
"epoch": 0.22913829994164558,
"grad_norm": 1.171875,
"learning_rate": 0.0004997420086365351,
"loss": 5.8379,
"mean_token_accuracy": 0.1387510970234871,
"num_tokens": 5123249.0,
"step": 2945
},
{
"entropy": 5.894250869750977,
"epoch": 0.22952732931336317,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997406800690966,
"loss": 5.8134,
"mean_token_accuracy": 0.138453159481287,
"num_tokens": 5131906.0,
"step": 2950
},
{
"entropy": 5.863798522949219,
"epoch": 0.22991635868508073,
"grad_norm": 1.2109375,
"learning_rate": 0.0004997393480915758,
"loss": 5.8003,
"mean_token_accuracy": 0.146034524589777,
"num_tokens": 5140740.0,
"step": 2955
},
{
"entropy": 5.900701189041138,
"epoch": 0.2303053880567983,
"grad_norm": 1.140625,
"learning_rate": 0.0004997380127039931,
"loss": 5.9917,
"mean_token_accuracy": 0.13965021297335625,
"num_tokens": 5149748.0,
"step": 2960
},
{
"entropy": 5.876333379745484,
"epoch": 0.23069441742851585,
"grad_norm": 1.140625,
"learning_rate": 0.0004997366739063688,
"loss": 5.719,
"mean_token_accuracy": 0.14996112510561943,
"num_tokens": 5158475.0,
"step": 2965
},
{
"entropy": 5.8662495613098145,
"epoch": 0.2310834468002334,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997353316987231,
"loss": 5.913,
"mean_token_accuracy": 0.13812702000141144,
"num_tokens": 5166593.0,
"step": 2970
},
{
"entropy": 5.93091893196106,
"epoch": 0.231472476171951,
"grad_norm": 1.203125,
"learning_rate": 0.0004997339860810765,
"loss": 5.8913,
"mean_token_accuracy": 0.14304886758327484,
"num_tokens": 5175557.0,
"step": 2975
},
{
"entropy": 5.873473691940307,
"epoch": 0.23186150554366855,
"grad_norm": 1.140625,
"learning_rate": 0.0004997326370534495,
"loss": 5.7973,
"mean_token_accuracy": 0.14339174702763557,
"num_tokens": 5184140.0,
"step": 2980
},
{
"entropy": 5.962021112442017,
"epoch": 0.2322505349153861,
"grad_norm": 1.1953125,
"learning_rate": 0.0004997312846158622,
"loss": 5.8895,
"mean_token_accuracy": 0.13540481626987458,
"num_tokens": 5192347.0,
"step": 2985
},
{
"entropy": 5.857995939254761,
"epoch": 0.23263956428710367,
"grad_norm": 1.203125,
"learning_rate": 0.0004997299287683355,
"loss": 5.8516,
"mean_token_accuracy": 0.1456963911652565,
"num_tokens": 5201327.0,
"step": 2990
},
{
"entropy": 5.914484262466431,
"epoch": 0.23302859365882125,
"grad_norm": 1.2421875,
"learning_rate": 0.0004997285695108898,
"loss": 5.9333,
"mean_token_accuracy": 0.14006564170122146,
"num_tokens": 5209873.0,
"step": 2995
},
{
"entropy": 5.987733602523804,
"epoch": 0.2334176230305388,
"grad_norm": 1.21875,
"learning_rate": 0.0004997272068435458,
"loss": 5.8368,
"mean_token_accuracy": 0.14314770549535752,
"num_tokens": 5219176.0,
"step": 3000
},
{
"epoch": 0.2334176230305388,
"eval_entropy": 5.733757502672417,
"eval_loss": 5.909061431884766,
"eval_mean_token_accuracy": 0.1476683286819123,
"eval_num_tokens": 5219176.0,
"eval_runtime": 28.6428,
"eval_samples_per_second": 1450.906,
"eval_steps_per_second": 181.372,
"step": 3000
},
{
"entropy": 5.831014919281006,
"epoch": 0.23380665240225637,
"grad_norm": 1.265625,
"learning_rate": 0.000499725840766324,
"loss": 5.781,
"mean_token_accuracy": 0.15310309678316117,
"num_tokens": 5227310.0,
"step": 3005
},
{
"entropy": 5.916271114349366,
"epoch": 0.23419568177397393,
"grad_norm": 1.25,
"learning_rate": 0.0004997244712792453,
"loss": 5.871,
"mean_token_accuracy": 0.13849457129836082,
"num_tokens": 5235882.0,
"step": 3010
},
{
"entropy": 5.8643897533416744,
"epoch": 0.2345847111456915,
"grad_norm": 1.203125,
"learning_rate": 0.0004997230983823305,
"loss": 5.7895,
"mean_token_accuracy": 0.14129805639386178,
"num_tokens": 5244149.0,
"step": 3015
},
{
"entropy": 5.725203800201416,
"epoch": 0.23497374051740907,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997217220756003,
"loss": 5.6533,
"mean_token_accuracy": 0.14943938702344894,
"num_tokens": 5252473.0,
"step": 3020
},
{
"entropy": 5.831255388259888,
"epoch": 0.23536276988912663,
"grad_norm": 1.171875,
"learning_rate": 0.0004997203423590757,
"loss": 5.8125,
"mean_token_accuracy": 0.14606485813856124,
"num_tokens": 5260972.0,
"step": 3025
},
{
"entropy": 5.885420131683349,
"epoch": 0.2357517992608442,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997189592327775,
"loss": 5.8225,
"mean_token_accuracy": 0.13927970007061957,
"num_tokens": 5270513.0,
"step": 3030
},
{
"entropy": 5.882440185546875,
"epoch": 0.23614082863256175,
"grad_norm": 1.203125,
"learning_rate": 0.0004997175726967268,
"loss": 5.803,
"mean_token_accuracy": 0.1392829418182373,
"num_tokens": 5278433.0,
"step": 3035
},
{
"entropy": 5.83907732963562,
"epoch": 0.23652985800427934,
"grad_norm": 1.2421875,
"learning_rate": 0.0004997161827509447,
"loss": 5.7977,
"mean_token_accuracy": 0.13926001712679864,
"num_tokens": 5287389.0,
"step": 3040
},
{
"entropy": 5.8517955303192135,
"epoch": 0.2369188873759969,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997147893954521,
"loss": 5.9115,
"mean_token_accuracy": 0.1402236580848694,
"num_tokens": 5296220.0,
"step": 3045
},
{
"entropy": 5.955138111114502,
"epoch": 0.23730791674771445,
"grad_norm": 1.203125,
"learning_rate": 0.0004997133926302702,
"loss": 5.8328,
"mean_token_accuracy": 0.14494856148958207,
"num_tokens": 5304152.0,
"step": 3050
},
{
"entropy": 5.933644962310791,
"epoch": 0.237696946119432,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997119924554204,
"loss": 5.8697,
"mean_token_accuracy": 0.14269917458295822,
"num_tokens": 5312374.0,
"step": 3055
},
{
"entropy": 5.800216960906982,
"epoch": 0.23808597549114957,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997105888709236,
"loss": 5.7876,
"mean_token_accuracy": 0.14500768929719926,
"num_tokens": 5320696.0,
"step": 3060
},
{
"entropy": 5.906084728240967,
"epoch": 0.23847500486286716,
"grad_norm": 1.2578125,
"learning_rate": 0.0004997091818768015,
"loss": 5.927,
"mean_token_accuracy": 0.13957108333706855,
"num_tokens": 5328754.0,
"step": 3065
},
{
"entropy": 5.924434232711792,
"epoch": 0.23886403423458472,
"grad_norm": 1.140625,
"learning_rate": 0.000499707771473075,
"loss": 5.878,
"mean_token_accuracy": 0.1405944973230362,
"num_tokens": 5337553.0,
"step": 3070
},
{
"entropy": 5.905769205093383,
"epoch": 0.23925306360630227,
"grad_norm": 1.2578125,
"learning_rate": 0.0004997063576597659,
"loss": 5.809,
"mean_token_accuracy": 0.14863870143890381,
"num_tokens": 5346063.0,
"step": 3075
},
{
"entropy": 5.871305704116821,
"epoch": 0.23964209297801983,
"grad_norm": 1.28125,
"learning_rate": 0.0004997049404368954,
"loss": 5.8745,
"mean_token_accuracy": 0.14547610208392142,
"num_tokens": 5354187.0,
"step": 3080
},
{
"entropy": 5.925069046020508,
"epoch": 0.2400311223497374,
"grad_norm": 1.3046875,
"learning_rate": 0.000499703519804485,
"loss": 5.9096,
"mean_token_accuracy": 0.1328617438673973,
"num_tokens": 5362819.0,
"step": 3085
},
{
"entropy": 5.897925138473511,
"epoch": 0.24042015172145498,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997020957625564,
"loss": 5.7919,
"mean_token_accuracy": 0.15332378596067428,
"num_tokens": 5371249.0,
"step": 3090
},
{
"entropy": 5.877030849456787,
"epoch": 0.24080918109317254,
"grad_norm": 1.140625,
"learning_rate": 0.0004997006683111312,
"loss": 5.7994,
"mean_token_accuracy": 0.1433065228164196,
"num_tokens": 5379284.0,
"step": 3095
},
{
"entropy": 5.884299421310425,
"epoch": 0.2411982104648901,
"grad_norm": 1.1328125,
"learning_rate": 0.000499699237450231,
"loss": 5.8996,
"mean_token_accuracy": 0.14278156086802482,
"num_tokens": 5388541.0,
"step": 3100
},
{
"entropy": 5.94818902015686,
"epoch": 0.24158723983660765,
"grad_norm": 1.125,
"learning_rate": 0.0004996978031798774,
"loss": 5.9485,
"mean_token_accuracy": 0.13989101499319076,
"num_tokens": 5397226.0,
"step": 3105
},
{
"entropy": 5.939749002456665,
"epoch": 0.24197626920832524,
"grad_norm": 1.2265625,
"learning_rate": 0.0004996963655000924,
"loss": 5.8573,
"mean_token_accuracy": 0.1397051528096199,
"num_tokens": 5405579.0,
"step": 3110
},
{
"entropy": 5.892177295684815,
"epoch": 0.2423652985800428,
"grad_norm": 1.171875,
"learning_rate": 0.0004996949244108977,
"loss": 5.9838,
"mean_token_accuracy": 0.12947663366794587,
"num_tokens": 5414529.0,
"step": 3115
},
{
"entropy": 5.9776526927948,
"epoch": 0.24275432795176036,
"grad_norm": 1.15625,
"learning_rate": 0.000499693479912315,
"loss": 5.9389,
"mean_token_accuracy": 0.14825934171676636,
"num_tokens": 5423641.0,
"step": 3120
},
{
"entropy": 5.853601503372192,
"epoch": 0.24314335732347792,
"grad_norm": 1.1875,
"learning_rate": 0.0004996920320043666,
"loss": 5.8381,
"mean_token_accuracy": 0.1416511431336403,
"num_tokens": 5432558.0,
"step": 3125
},
{
"entropy": 5.866245222091675,
"epoch": 0.24353238669519547,
"grad_norm": 1.1875,
"learning_rate": 0.0004996905806870742,
"loss": 5.7816,
"mean_token_accuracy": 0.14246323853731155,
"num_tokens": 5441449.0,
"step": 3130
},
{
"entropy": 5.867586135864258,
"epoch": 0.24392141606691306,
"grad_norm": 1.015625,
"learning_rate": 0.0004996891259604598,
"loss": 5.7997,
"mean_token_accuracy": 0.14338409453630446,
"num_tokens": 5451142.0,
"step": 3135
},
{
"entropy": 5.896688175201416,
"epoch": 0.24431044543863062,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996876678245455,
"loss": 5.7613,
"mean_token_accuracy": 0.14048081785440444,
"num_tokens": 5459290.0,
"step": 3140
},
{
"entropy": 5.798839139938354,
"epoch": 0.24469947481034818,
"grad_norm": 1.296875,
"learning_rate": 0.0004996862062793536,
"loss": 5.7811,
"mean_token_accuracy": 0.1482220396399498,
"num_tokens": 5467776.0,
"step": 3145
},
{
"entropy": 5.7967747211456295,
"epoch": 0.24508850418206574,
"grad_norm": 1.140625,
"learning_rate": 0.0004996847413249061,
"loss": 5.6617,
"mean_token_accuracy": 0.14994954615831374,
"num_tokens": 5476005.0,
"step": 3150
},
{
"entropy": 5.8860608577728275,
"epoch": 0.24547753355378332,
"grad_norm": 1.1640625,
"learning_rate": 0.0004996832729612252,
"loss": 5.8461,
"mean_token_accuracy": 0.14408416971564292,
"num_tokens": 5484225.0,
"step": 3155
},
{
"entropy": 5.765978288650513,
"epoch": 0.24586656292550088,
"grad_norm": 1.125,
"learning_rate": 0.0004996818011883333,
"loss": 5.7774,
"mean_token_accuracy": 0.14943447411060334,
"num_tokens": 5493111.0,
"step": 3160
},
{
"entropy": 5.831310415267945,
"epoch": 0.24625559229721844,
"grad_norm": 1.2109375,
"learning_rate": 0.0004996803260062527,
"loss": 5.7948,
"mean_token_accuracy": 0.14830601513385772,
"num_tokens": 5501574.0,
"step": 3165
},
{
"entropy": 5.875056219100952,
"epoch": 0.246644621668936,
"grad_norm": 1.2578125,
"learning_rate": 0.0004996788474150057,
"loss": 5.8249,
"mean_token_accuracy": 0.1414250537753105,
"num_tokens": 5511136.0,
"step": 3170
},
{
"entropy": 5.807733631134033,
"epoch": 0.24703365104065356,
"grad_norm": 1.1640625,
"learning_rate": 0.0004996773654146149,
"loss": 5.8168,
"mean_token_accuracy": 0.14713762253522872,
"num_tokens": 5520324.0,
"step": 3175
},
{
"entropy": 5.874153709411621,
"epoch": 0.24742268041237114,
"grad_norm": 1.140625,
"learning_rate": 0.0004996758800051026,
"loss": 5.7661,
"mean_token_accuracy": 0.14694686979055405,
"num_tokens": 5528960.0,
"step": 3180
},
{
"entropy": 5.865899896621704,
"epoch": 0.2478117097840887,
"grad_norm": 1.4140625,
"learning_rate": 0.0004996743911864916,
"loss": 5.926,
"mean_token_accuracy": 0.14135217815637588,
"num_tokens": 5537391.0,
"step": 3185
},
{
"entropy": 5.90128002166748,
"epoch": 0.24820073915580626,
"grad_norm": 1.140625,
"learning_rate": 0.0004996728989588041,
"loss": 5.8662,
"mean_token_accuracy": 0.13981591016054154,
"num_tokens": 5546577.0,
"step": 3190
},
{
"entropy": 5.93365888595581,
"epoch": 0.24858976852752382,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996714033220632,
"loss": 5.8035,
"mean_token_accuracy": 0.1451164186000824,
"num_tokens": 5555247.0,
"step": 3195
},
{
"entropy": 5.79735369682312,
"epoch": 0.2489787978992414,
"grad_norm": 1.2734375,
"learning_rate": 0.0004996699042762911,
"loss": 5.7679,
"mean_token_accuracy": 0.14924910962581633,
"num_tokens": 5563598.0,
"step": 3200
},
{
"entropy": 5.854664468765259,
"epoch": 0.24936782727095896,
"grad_norm": 1.2421875,
"learning_rate": 0.0004996684018215111,
"loss": 5.7983,
"mean_token_accuracy": 0.14165054261684418,
"num_tokens": 5572734.0,
"step": 3205
},
{
"entropy": 5.876973295211792,
"epoch": 0.24975685664267652,
"grad_norm": 1.15625,
"learning_rate": 0.0004996668959577453,
"loss": 5.9106,
"mean_token_accuracy": 0.1385761357843876,
"num_tokens": 5581224.0,
"step": 3210
},
{
"entropy": 5.934681081771851,
"epoch": 0.2501458860143941,
"grad_norm": 1.140625,
"learning_rate": 0.0004996653866850173,
"loss": 5.9043,
"mean_token_accuracy": 0.13491922244429588,
"num_tokens": 5590138.0,
"step": 3215
},
{
"entropy": 5.842910432815552,
"epoch": 0.25053491538611167,
"grad_norm": 1.2578125,
"learning_rate": 0.0004996638740033495,
"loss": 5.837,
"mean_token_accuracy": 0.149862140417099,
"num_tokens": 5598654.0,
"step": 3220
},
{
"entropy": 5.754105043411255,
"epoch": 0.2509239447578292,
"grad_norm": 1.2734375,
"learning_rate": 0.0004996623579127651,
"loss": 5.7791,
"mean_token_accuracy": 0.14800113886594773,
"num_tokens": 5607404.0,
"step": 3225
},
{
"entropy": 5.850812196731567,
"epoch": 0.2513129741295468,
"grad_norm": 1.3046875,
"learning_rate": 0.0004996608384132868,
"loss": 5.8029,
"mean_token_accuracy": 0.13932496458292007,
"num_tokens": 5617133.0,
"step": 3230
},
{
"entropy": 5.841469144821167,
"epoch": 0.25170200350126437,
"grad_norm": 1.1875,
"learning_rate": 0.000499659315504938,
"loss": 5.8294,
"mean_token_accuracy": 0.13994766473770143,
"num_tokens": 5625389.0,
"step": 3235
},
{
"entropy": 5.826369094848633,
"epoch": 0.2520910328729819,
"grad_norm": 1.203125,
"learning_rate": 0.0004996577891877417,
"loss": 5.7015,
"mean_token_accuracy": 0.15945661962032318,
"num_tokens": 5633985.0,
"step": 3240
},
{
"entropy": 5.821769857406617,
"epoch": 0.2524800622446995,
"grad_norm": 1.2890625,
"learning_rate": 0.0004996562594617209,
"loss": 5.8316,
"mean_token_accuracy": 0.13962800428271294,
"num_tokens": 5642524.0,
"step": 3245
},
{
"entropy": 5.852222156524658,
"epoch": 0.252869091616417,
"grad_norm": 1.3046875,
"learning_rate": 0.0004996547263268991,
"loss": 5.7793,
"mean_token_accuracy": 0.1448754347860813,
"num_tokens": 5651788.0,
"step": 3250
},
{
"entropy": 5.821958208084107,
"epoch": 0.2532581209881346,
"grad_norm": 1.15625,
"learning_rate": 0.0004996531897832993,
"loss": 5.8159,
"mean_token_accuracy": 0.1427738018333912,
"num_tokens": 5660757.0,
"step": 3255
},
{
"entropy": 5.984094285964966,
"epoch": 0.2536471503598522,
"grad_norm": 1.1484375,
"learning_rate": 0.000499651649830945,
"loss": 5.9304,
"mean_token_accuracy": 0.14311277642846107,
"num_tokens": 5669760.0,
"step": 3260
},
{
"entropy": 5.813133239746094,
"epoch": 0.2540361797315697,
"grad_norm": 1.1484375,
"learning_rate": 0.0004996501064698593,
"loss": 5.8492,
"mean_token_accuracy": 0.14234110862016677,
"num_tokens": 5678313.0,
"step": 3265
},
{
"entropy": 5.870118856430054,
"epoch": 0.2544252091032873,
"grad_norm": 1.203125,
"learning_rate": 0.000499648559700066,
"loss": 5.7668,
"mean_token_accuracy": 0.15043241605162622,
"num_tokens": 5686942.0,
"step": 3270
},
{
"entropy": 5.792041444778443,
"epoch": 0.25481423847500484,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996470095215884,
"loss": 5.7555,
"mean_token_accuracy": 0.14962123110890388,
"num_tokens": 5695612.0,
"step": 3275
},
{
"entropy": 5.900144004821778,
"epoch": 0.2552032678467224,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996454559344498,
"loss": 5.9529,
"mean_token_accuracy": 0.13228990584611894,
"num_tokens": 5705067.0,
"step": 3280
},
{
"entropy": 5.906079196929932,
"epoch": 0.25559229721844,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996438989386741,
"loss": 5.8355,
"mean_token_accuracy": 0.14319185838103293,
"num_tokens": 5714092.0,
"step": 3285
},
{
"entropy": 5.8403573513031,
"epoch": 0.25598132659015754,
"grad_norm": 1.140625,
"learning_rate": 0.0004996423385342847,
"loss": 5.8028,
"mean_token_accuracy": 0.14620732218027116,
"num_tokens": 5722638.0,
"step": 3290
},
{
"entropy": 5.824834823608398,
"epoch": 0.25637035596187513,
"grad_norm": 1.1484375,
"learning_rate": 0.0004996407747213055,
"loss": 5.804,
"mean_token_accuracy": 0.14224600344896315,
"num_tokens": 5731511.0,
"step": 3295
},
{
"entropy": 5.861731052398682,
"epoch": 0.25675938533359266,
"grad_norm": 1.1171875,
"learning_rate": 0.00049963920749976,
"loss": 5.8202,
"mean_token_accuracy": 0.14186736792325974,
"num_tokens": 5740950.0,
"step": 3300
},
{
"entropy": 5.807591533660888,
"epoch": 0.25714841470531025,
"grad_norm": 1.1953125,
"learning_rate": 0.0004996376368696721,
"loss": 5.7591,
"mean_token_accuracy": 0.14281872063875198,
"num_tokens": 5749163.0,
"step": 3305
},
{
"entropy": 5.8925347328186035,
"epoch": 0.25753744407702783,
"grad_norm": 1.1796875,
"learning_rate": 0.0004996360628310656,
"loss": 5.8173,
"mean_token_accuracy": 0.14276841878890992,
"num_tokens": 5758321.0,
"step": 3310
},
{
"entropy": 5.9321657657623295,
"epoch": 0.25792647344874536,
"grad_norm": 1.1953125,
"learning_rate": 0.0004996344853839644,
"loss": 5.8682,
"mean_token_accuracy": 0.14533354043960572,
"num_tokens": 5767277.0,
"step": 3315
},
{
"entropy": 5.820040512084961,
"epoch": 0.25831550282046295,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996329045283924,
"loss": 5.7605,
"mean_token_accuracy": 0.150101937353611,
"num_tokens": 5775694.0,
"step": 3320
},
{
"entropy": 5.734701347351074,
"epoch": 0.25870453219218054,
"grad_norm": 1.09375,
"learning_rate": 0.0004996313202643737,
"loss": 5.6773,
"mean_token_accuracy": 0.15130480825901033,
"num_tokens": 5783932.0,
"step": 3325
},
{
"entropy": 5.83421835899353,
"epoch": 0.25909356156389807,
"grad_norm": 1.1875,
"learning_rate": 0.0004996297325919321,
"loss": 5.8049,
"mean_token_accuracy": 0.14608783572912215,
"num_tokens": 5792100.0,
"step": 3330
},
{
"entropy": 5.818102645874023,
"epoch": 0.25948259093561565,
"grad_norm": 1.1484375,
"learning_rate": 0.0004996281415110919,
"loss": 5.7919,
"mean_token_accuracy": 0.1446779452264309,
"num_tokens": 5801481.0,
"step": 3335
},
{
"entropy": 5.843531894683838,
"epoch": 0.2598716203073332,
"grad_norm": 1.125,
"learning_rate": 0.0004996265470218773,
"loss": 5.8885,
"mean_token_accuracy": 0.13893020674586296,
"num_tokens": 5810053.0,
"step": 3340
},
{
"entropy": 5.82769284248352,
"epoch": 0.26026064967905077,
"grad_norm": 1.109375,
"learning_rate": 0.0004996249491243122,
"loss": 5.6783,
"mean_token_accuracy": 0.15309639424085617,
"num_tokens": 5818454.0,
"step": 3345
},
{
"entropy": 5.812572193145752,
"epoch": 0.26064967905076836,
"grad_norm": 1.140625,
"learning_rate": 0.0004996233478184211,
"loss": 5.8512,
"mean_token_accuracy": 0.13824255466461183,
"num_tokens": 5827013.0,
"step": 3350
},
{
"entropy": 5.805325746536255,
"epoch": 0.2610387084224859,
"grad_norm": 1.203125,
"learning_rate": 0.0004996217431042282,
"loss": 5.8482,
"mean_token_accuracy": 0.1430810771882534,
"num_tokens": 5835862.0,
"step": 3355
},
{
"entropy": 5.7693298816680905,
"epoch": 0.2614277377942035,
"grad_norm": 1.0546875,
"learning_rate": 0.000499620134981758,
"loss": 5.6191,
"mean_token_accuracy": 0.15031312257051468,
"num_tokens": 5843827.0,
"step": 3360
},
{
"entropy": 5.80864987373352,
"epoch": 0.261816767165921,
"grad_norm": 1.109375,
"learning_rate": 0.0004996185234510346,
"loss": 5.8272,
"mean_token_accuracy": 0.13573512434959412,
"num_tokens": 5852677.0,
"step": 3365
},
{
"entropy": 5.898306369781494,
"epoch": 0.2622057965376386,
"grad_norm": 1.2265625,
"learning_rate": 0.0004996169085120828,
"loss": 5.7732,
"mean_token_accuracy": 0.1503363937139511,
"num_tokens": 5861686.0,
"step": 3370
},
{
"entropy": 5.7120969772338865,
"epoch": 0.2625948259093562,
"grad_norm": 1.109375,
"learning_rate": 0.0004996152901649268,
"loss": 5.7407,
"mean_token_accuracy": 0.14560485854744912,
"num_tokens": 5870690.0,
"step": 3375
},
{
"entropy": 5.79764723777771,
"epoch": 0.2629838552810737,
"grad_norm": 1.15625,
"learning_rate": 0.0004996136684095913,
"loss": 5.8249,
"mean_token_accuracy": 0.14097655266523362,
"num_tokens": 5879804.0,
"step": 3380
},
{
"entropy": 5.94910888671875,
"epoch": 0.2633728846527913,
"grad_norm": 1.3515625,
"learning_rate": 0.0004996120432461009,
"loss": 5.8967,
"mean_token_accuracy": 0.13631573617458342,
"num_tokens": 5888781.0,
"step": 3385
},
{
"entropy": 5.912099456787109,
"epoch": 0.2637619140245088,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996104146744804,
"loss": 5.808,
"mean_token_accuracy": 0.14571621865034104,
"num_tokens": 5897439.0,
"step": 3390
},
{
"entropy": 5.829486560821533,
"epoch": 0.2641509433962264,
"grad_norm": 1.21875,
"learning_rate": 0.0004996087826947541,
"loss": 5.7533,
"mean_token_accuracy": 0.1568281203508377,
"num_tokens": 5905471.0,
"step": 3395
},
{
"entropy": 5.755701017379761,
"epoch": 0.264539972767944,
"grad_norm": 1.25,
"learning_rate": 0.0004996071473069471,
"loss": 5.8004,
"mean_token_accuracy": 0.14211232513189315,
"num_tokens": 5913795.0,
"step": 3400
},
{
"entropy": 5.936715173721313,
"epoch": 0.26492900213966153,
"grad_norm": 1.2421875,
"learning_rate": 0.0004996055085110842,
"loss": 5.8668,
"mean_token_accuracy": 0.1416642837226391,
"num_tokens": 5922119.0,
"step": 3405
},
{
"entropy": 5.782314729690552,
"epoch": 0.2653180315113791,
"grad_norm": 1.2578125,
"learning_rate": 0.0004996038663071902,
"loss": 5.7469,
"mean_token_accuracy": 0.14390047043561935,
"num_tokens": 5930759.0,
"step": 3410
},
{
"entropy": 5.816781759262085,
"epoch": 0.26570706088309665,
"grad_norm": 1.2265625,
"learning_rate": 0.0004996022206952901,
"loss": 5.7809,
"mean_token_accuracy": 0.1507711961865425,
"num_tokens": 5939830.0,
"step": 3415
},
{
"entropy": 5.91773362159729,
"epoch": 0.26609609025481423,
"grad_norm": 1.1875,
"learning_rate": 0.0004996005716754086,
"loss": 5.7921,
"mean_token_accuracy": 0.143372742831707,
"num_tokens": 5947737.0,
"step": 3420
},
{
"entropy": 5.702203607559204,
"epoch": 0.2664851196265318,
"grad_norm": 1.234375,
"learning_rate": 0.000499598919247571,
"loss": 5.7436,
"mean_token_accuracy": 0.1547391414642334,
"num_tokens": 5955905.0,
"step": 3425
},
{
"entropy": 5.761687517166138,
"epoch": 0.26687414899824935,
"grad_norm": 1.265625,
"learning_rate": 0.0004995972634118023,
"loss": 5.7991,
"mean_token_accuracy": 0.14444849416613578,
"num_tokens": 5965223.0,
"step": 3430
},
{
"entropy": 5.886661386489868,
"epoch": 0.26726317836996694,
"grad_norm": 1.1796875,
"learning_rate": 0.0004995956041681277,
"loss": 5.8049,
"mean_token_accuracy": 0.1485450528562069,
"num_tokens": 5973615.0,
"step": 3435
},
{
"entropy": 5.813414812088013,
"epoch": 0.2676522077416845,
"grad_norm": 1.2265625,
"learning_rate": 0.0004995939415165722,
"loss": 5.7458,
"mean_token_accuracy": 0.15244108885526658,
"num_tokens": 5981600.0,
"step": 3440
},
{
"entropy": 5.789648532867432,
"epoch": 0.26804123711340205,
"grad_norm": 1.1875,
"learning_rate": 0.0004995922754571611,
"loss": 5.7446,
"mean_token_accuracy": 0.1445736438035965,
"num_tokens": 5989928.0,
"step": 3445
},
{
"entropy": 5.745592164993286,
"epoch": 0.26843026648511964,
"grad_norm": 1.1953125,
"learning_rate": 0.0004995906059899197,
"loss": 5.8116,
"mean_token_accuracy": 0.14425835460424424,
"num_tokens": 5998704.0,
"step": 3450
},
{
"entropy": 5.887583017349243,
"epoch": 0.26881929585683717,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995889331148733,
"loss": 5.7879,
"mean_token_accuracy": 0.14587609022855758,
"num_tokens": 6007476.0,
"step": 3455
},
{
"entropy": 5.881060028076172,
"epoch": 0.26920832522855476,
"grad_norm": 1.171875,
"learning_rate": 0.0004995872568320474,
"loss": 5.8628,
"mean_token_accuracy": 0.14118837639689447,
"num_tokens": 6016324.0,
"step": 3460
},
{
"entropy": 5.846192836761475,
"epoch": 0.26959735460027234,
"grad_norm": 1.21875,
"learning_rate": 0.0004995855771414673,
"loss": 5.7784,
"mean_token_accuracy": 0.14671637415885924,
"num_tokens": 6024709.0,
"step": 3465
},
{
"entropy": 5.764834785461426,
"epoch": 0.2699863839719899,
"grad_norm": 1.203125,
"learning_rate": 0.0004995838940431585,
"loss": 5.6378,
"mean_token_accuracy": 0.1520429238677025,
"num_tokens": 6033131.0,
"step": 3470
},
{
"entropy": 5.751634168624878,
"epoch": 0.27037541334370746,
"grad_norm": 1.1484375,
"learning_rate": 0.0004995822075371466,
"loss": 5.8405,
"mean_token_accuracy": 0.14594754576683044,
"num_tokens": 6042310.0,
"step": 3475
},
{
"entropy": 5.842454528808593,
"epoch": 0.270764442715425,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995805176234571,
"loss": 5.6754,
"mean_token_accuracy": 0.15385707169771196,
"num_tokens": 6050617.0,
"step": 3480
},
{
"entropy": 5.82150673866272,
"epoch": 0.2711534720871426,
"grad_norm": 1.125,
"learning_rate": 0.0004995788243021158,
"loss": 5.8341,
"mean_token_accuracy": 0.14465887770056723,
"num_tokens": 6060090.0,
"step": 3485
},
{
"entropy": 5.870070219039917,
"epoch": 0.27154250145886016,
"grad_norm": 1.1796875,
"learning_rate": 0.0004995771275731483,
"loss": 5.7816,
"mean_token_accuracy": 0.14922453165054322,
"num_tokens": 6068646.0,
"step": 3490
},
{
"entropy": 5.840321493148804,
"epoch": 0.2719315308305777,
"grad_norm": 1.3359375,
"learning_rate": 0.0004995754274365802,
"loss": 5.8403,
"mean_token_accuracy": 0.14855221211910247,
"num_tokens": 6077896.0,
"step": 3495
},
{
"entropy": 5.796200132369995,
"epoch": 0.2723205602022953,
"grad_norm": 1.15625,
"learning_rate": 0.0004995737238924376,
"loss": 5.7522,
"mean_token_accuracy": 0.14376320838928222,
"num_tokens": 6087189.0,
"step": 3500
},
{
"entropy": 5.823796129226684,
"epoch": 0.2727095895740128,
"grad_norm": 1.2265625,
"learning_rate": 0.0004995720169407461,
"loss": 5.7489,
"mean_token_accuracy": 0.1495093137025833,
"num_tokens": 6095394.0,
"step": 3505
},
{
"entropy": 5.836558055877686,
"epoch": 0.2730986189457304,
"grad_norm": 1.28125,
"learning_rate": 0.0004995703065815317,
"loss": 5.7164,
"mean_token_accuracy": 0.1526440754532814,
"num_tokens": 6104135.0,
"step": 3510
},
{
"entropy": 5.823418998718262,
"epoch": 0.273487648317448,
"grad_norm": 1.28125,
"learning_rate": 0.0004995685928148203,
"loss": 5.8308,
"mean_token_accuracy": 0.1389838382601738,
"num_tokens": 6112160.0,
"step": 3515
},
{
"entropy": 5.857429122924804,
"epoch": 0.2738766776891655,
"grad_norm": 1.203125,
"learning_rate": 0.000499566875640638,
"loss": 5.8379,
"mean_token_accuracy": 0.15000538304448127,
"num_tokens": 6120575.0,
"step": 3520
},
{
"entropy": 5.860440254211426,
"epoch": 0.2742657070608831,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995651550590108,
"loss": 5.8404,
"mean_token_accuracy": 0.1447710335254669,
"num_tokens": 6129367.0,
"step": 3525
},
{
"entropy": 5.855791473388672,
"epoch": 0.2746547364326007,
"grad_norm": 1.2421875,
"learning_rate": 0.0004995634310699647,
"loss": 5.8259,
"mean_token_accuracy": 0.14781430512666702,
"num_tokens": 6137440.0,
"step": 3530
},
{
"entropy": 5.7965374946594235,
"epoch": 0.2750437658043182,
"grad_norm": 1.109375,
"learning_rate": 0.0004995617036735261,
"loss": 5.7402,
"mean_token_accuracy": 0.14770583286881447,
"num_tokens": 6146770.0,
"step": 3535
},
{
"entropy": 5.904029750823975,
"epoch": 0.2754327951760358,
"grad_norm": 1.1953125,
"learning_rate": 0.0004995599728697211,
"loss": 5.8318,
"mean_token_accuracy": 0.14512663930654526,
"num_tokens": 6155994.0,
"step": 3540
},
{
"entropy": 5.800452709197998,
"epoch": 0.27582182454775334,
"grad_norm": 1.2265625,
"learning_rate": 0.0004995582386585759,
"loss": 5.7515,
"mean_token_accuracy": 0.14241246953606607,
"num_tokens": 6163941.0,
"step": 3545
},
{
"entropy": 5.767463302612304,
"epoch": 0.2762108539194709,
"grad_norm": 1.234375,
"learning_rate": 0.0004995565010401167,
"loss": 5.7941,
"mean_token_accuracy": 0.14686818718910216,
"num_tokens": 6172681.0,
"step": 3550
},
{
"entropy": 5.897072124481201,
"epoch": 0.2765998832911885,
"grad_norm": 1.171875,
"learning_rate": 0.0004995547600143702,
"loss": 5.7758,
"mean_token_accuracy": 0.1455358862876892,
"num_tokens": 6181449.0,
"step": 3555
},
{
"entropy": 5.82000322341919,
"epoch": 0.27698891266290604,
"grad_norm": 1.1640625,
"learning_rate": 0.0004995530155813625,
"loss": 5.7555,
"mean_token_accuracy": 0.1465877875685692,
"num_tokens": 6190275.0,
"step": 3560
},
{
"entropy": 5.805966186523437,
"epoch": 0.2773779420346236,
"grad_norm": 1.1484375,
"learning_rate": 0.0004995512677411203,
"loss": 5.7792,
"mean_token_accuracy": 0.14627898782491683,
"num_tokens": 6198521.0,
"step": 3565
},
{
"entropy": 5.8195501327514645,
"epoch": 0.27776697140634116,
"grad_norm": 1.15625,
"learning_rate": 0.0004995495164936698,
"loss": 5.8445,
"mean_token_accuracy": 0.1397809185087681,
"num_tokens": 6207318.0,
"step": 3570
},
{
"entropy": 5.87066617012024,
"epoch": 0.27815600077805874,
"grad_norm": 1.15625,
"learning_rate": 0.0004995477618390381,
"loss": 5.7298,
"mean_token_accuracy": 0.14824186712503434,
"num_tokens": 6216693.0,
"step": 3575
},
{
"entropy": 5.805923700332642,
"epoch": 0.27854503014977633,
"grad_norm": 1.2421875,
"learning_rate": 0.0004995460037772513,
"loss": 5.7919,
"mean_token_accuracy": 0.14640319496393203,
"num_tokens": 6225207.0,
"step": 3580
},
{
"entropy": 5.809096050262451,
"epoch": 0.27893405952149386,
"grad_norm": 1.3515625,
"learning_rate": 0.0004995442423083365,
"loss": 5.7325,
"mean_token_accuracy": 0.15586716383695604,
"num_tokens": 6232653.0,
"step": 3585
},
{
"entropy": 5.748783349990845,
"epoch": 0.27932308889321145,
"grad_norm": 1.1953125,
"learning_rate": 0.0004995424774323201,
"loss": 5.7295,
"mean_token_accuracy": 0.14759937971830367,
"num_tokens": 6241093.0,
"step": 3590
},
{
"entropy": 5.7409566879272464,
"epoch": 0.279712118264929,
"grad_norm": 1.2109375,
"learning_rate": 0.000499540709149229,
"loss": 5.6149,
"mean_token_accuracy": 0.15389250069856644,
"num_tokens": 6249144.0,
"step": 3595
},
{
"entropy": 5.788210344314575,
"epoch": 0.28010114763664656,
"grad_norm": 1.203125,
"learning_rate": 0.0004995389374590901,
"loss": 5.7058,
"mean_token_accuracy": 0.14813510403037072,
"num_tokens": 6257622.0,
"step": 3600
},
{
"entropy": 5.778064393997193,
"epoch": 0.28049017700836415,
"grad_norm": 1.1875,
"learning_rate": 0.0004995371623619301,
"loss": 5.7337,
"mean_token_accuracy": 0.14399215206503868,
"num_tokens": 6266435.0,
"step": 3605
},
{
"entropy": 5.855746650695801,
"epoch": 0.2808792063800817,
"grad_norm": 1.1953125,
"learning_rate": 0.000499535383857776,
"loss": 5.8548,
"mean_token_accuracy": 0.14212341010570526,
"num_tokens": 6275053.0,
"step": 3610
},
{
"entropy": 5.812265682220459,
"epoch": 0.28126823575179927,
"grad_norm": 1.2109375,
"learning_rate": 0.000499533601946655,
"loss": 5.7463,
"mean_token_accuracy": 0.1406642623245716,
"num_tokens": 6283240.0,
"step": 3615
},
{
"entropy": 5.72180495262146,
"epoch": 0.2816572651235168,
"grad_norm": 1.3203125,
"learning_rate": 0.0004995318166285938,
"loss": 5.6995,
"mean_token_accuracy": 0.15216349810361862,
"num_tokens": 6292133.0,
"step": 3620
},
{
"entropy": 5.851174068450928,
"epoch": 0.2820462944952344,
"grad_norm": 1.1953125,
"learning_rate": 0.0004995300279036199,
"loss": 5.8047,
"mean_token_accuracy": 0.14933415353298188,
"num_tokens": 6300954.0,
"step": 3625
},
{
"entropy": 5.751334381103516,
"epoch": 0.28243532386695197,
"grad_norm": 1.234375,
"learning_rate": 0.00049952823577176,
"loss": 5.6909,
"mean_token_accuracy": 0.15321999713778495,
"num_tokens": 6309637.0,
"step": 3630
},
{
"entropy": 5.790732383728027,
"epoch": 0.2828243532386695,
"grad_norm": 1.234375,
"learning_rate": 0.0004995264402330416,
"loss": 5.7623,
"mean_token_accuracy": 0.15102692544460297,
"num_tokens": 6318539.0,
"step": 3635
},
{
"entropy": 5.823702001571656,
"epoch": 0.2832133826103871,
"grad_norm": 1.21875,
"learning_rate": 0.0004995246412874919,
"loss": 5.686,
"mean_token_accuracy": 0.1494807243347168,
"num_tokens": 6327109.0,
"step": 3640
},
{
"entropy": 5.704149532318115,
"epoch": 0.2836024119821047,
"grad_norm": 1.21875,
"learning_rate": 0.000499522838935138,
"loss": 5.6806,
"mean_token_accuracy": 0.1496465563774109,
"num_tokens": 6335555.0,
"step": 3645
},
{
"entropy": 5.860112619400025,
"epoch": 0.2839914413538222,
"grad_norm": 1.15625,
"learning_rate": 0.0004995210331760074,
"loss": 5.8385,
"mean_token_accuracy": 0.1461038865149021,
"num_tokens": 6344716.0,
"step": 3650
},
{
"entropy": 5.746112537384033,
"epoch": 0.2843804707255398,
"grad_norm": 1.2421875,
"learning_rate": 0.0004995192240101275,
"loss": 5.7075,
"mean_token_accuracy": 0.15382865816354752,
"num_tokens": 6353618.0,
"step": 3655
},
{
"entropy": 5.873103380203247,
"epoch": 0.2847695000972573,
"grad_norm": 1.1953125,
"learning_rate": 0.0004995174114375258,
"loss": 5.8181,
"mean_token_accuracy": 0.14162709265947343,
"num_tokens": 6362039.0,
"step": 3660
},
{
"entropy": 5.856787061691284,
"epoch": 0.2851585294689749,
"grad_norm": 1.1796875,
"learning_rate": 0.0004995155954582297,
"loss": 5.7471,
"mean_token_accuracy": 0.14587011486291884,
"num_tokens": 6370331.0,
"step": 3665
},
{
"entropy": 5.742863416671753,
"epoch": 0.2855475588406925,
"grad_norm": 1.2265625,
"learning_rate": 0.0004995137760722668,
"loss": 5.7323,
"mean_token_accuracy": 0.14978418201208116,
"num_tokens": 6378943.0,
"step": 3670
},
{
"entropy": 5.848138999938965,
"epoch": 0.28593658821241,
"grad_norm": 1.203125,
"learning_rate": 0.0004995119532796646,
"loss": 5.7812,
"mean_token_accuracy": 0.13732778877019883,
"num_tokens": 6387745.0,
"step": 3675
},
{
"entropy": 5.648930788040161,
"epoch": 0.2863256175841276,
"grad_norm": 1.203125,
"learning_rate": 0.000499510127080451,
"loss": 5.6492,
"mean_token_accuracy": 0.14844874367117883,
"num_tokens": 6396915.0,
"step": 3680
},
{
"entropy": 5.686594676971436,
"epoch": 0.28671464695584514,
"grad_norm": 1.1796875,
"learning_rate": 0.0004995082974746535,
"loss": 5.6944,
"mean_token_accuracy": 0.15163496807217597,
"num_tokens": 6405453.0,
"step": 3685
},
{
"entropy": 5.84037594795227,
"epoch": 0.28710367632756273,
"grad_norm": 1.2578125,
"learning_rate": 0.0004995064644622999,
"loss": 5.7669,
"mean_token_accuracy": 0.14520054906606675,
"num_tokens": 6413693.0,
"step": 3690
},
{
"entropy": 5.780052042007446,
"epoch": 0.2874927056992803,
"grad_norm": 1.2734375,
"learning_rate": 0.0004995046280434181,
"loss": 5.7341,
"mean_token_accuracy": 0.14460751190781593,
"num_tokens": 6422738.0,
"step": 3695
},
{
"entropy": 5.742298173904419,
"epoch": 0.28788173507099785,
"grad_norm": 1.25,
"learning_rate": 0.0004995027882180358,
"loss": 5.6937,
"mean_token_accuracy": 0.15133106857538223,
"num_tokens": 6431743.0,
"step": 3700
},
{
"entropy": 5.809255313873291,
"epoch": 0.28827076444271543,
"grad_norm": 1.2109375,
"learning_rate": 0.0004995009449861812,
"loss": 5.7476,
"mean_token_accuracy": 0.1464826263487339,
"num_tokens": 6440586.0,
"step": 3705
},
{
"entropy": 5.792600870132446,
"epoch": 0.28865979381443296,
"grad_norm": 1.2890625,
"learning_rate": 0.000499499098347882,
"loss": 5.7945,
"mean_token_accuracy": 0.14538254216313362,
"num_tokens": 6449644.0,
"step": 3710
},
{
"entropy": 5.732555389404297,
"epoch": 0.28904882318615055,
"grad_norm": 1.125,
"learning_rate": 0.0004994972483031662,
"loss": 5.6924,
"mean_token_accuracy": 0.15000784248113633,
"num_tokens": 6458752.0,
"step": 3715
},
{
"entropy": 5.722676849365234,
"epoch": 0.28943785255786814,
"grad_norm": 1.1328125,
"learning_rate": 0.000499495394852062,
"loss": 5.7457,
"mean_token_accuracy": 0.15378984808921814,
"num_tokens": 6468333.0,
"step": 3720
},
{
"entropy": 5.651209449768066,
"epoch": 0.28982688192958567,
"grad_norm": 1.2421875,
"learning_rate": 0.0004994935379945977,
"loss": 5.6043,
"mean_token_accuracy": 0.15003781169652938,
"num_tokens": 6477406.0,
"step": 3725
},
{
"entropy": 5.86032395362854,
"epoch": 0.29021591130130325,
"grad_norm": 1.3671875,
"learning_rate": 0.0004994916777308012,
"loss": 5.768,
"mean_token_accuracy": 0.14514341950416565,
"num_tokens": 6486399.0,
"step": 3730
},
{
"entropy": 5.791876649856567,
"epoch": 0.29060494067302084,
"grad_norm": 1.171875,
"learning_rate": 0.0004994898140607007,
"loss": 5.6616,
"mean_token_accuracy": 0.14688180163502693,
"num_tokens": 6494853.0,
"step": 3735
},
{
"entropy": 5.795904016494751,
"epoch": 0.29099397004473837,
"grad_norm": 1.3125,
"learning_rate": 0.0004994879469843247,
"loss": 5.8281,
"mean_token_accuracy": 0.14508696421980857,
"num_tokens": 6503301.0,
"step": 3740
},
{
"entropy": 5.770627212524414,
"epoch": 0.29138299941645596,
"grad_norm": 1.078125,
"learning_rate": 0.0004994860765017013,
"loss": 5.7042,
"mean_token_accuracy": 0.14785689413547515,
"num_tokens": 6512685.0,
"step": 3745
},
{
"entropy": 5.811524724960327,
"epoch": 0.2917720287881735,
"grad_norm": 1.203125,
"learning_rate": 0.000499484202612859,
"loss": 5.83,
"mean_token_accuracy": 0.13926837593317032,
"num_tokens": 6520673.0,
"step": 3750
},
{
"entropy": 5.8250175476074215,
"epoch": 0.2921610581598911,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994823253178264,
"loss": 5.7345,
"mean_token_accuracy": 0.1433857038617134,
"num_tokens": 6529039.0,
"step": 3755
},
{
"entropy": 5.804865884780884,
"epoch": 0.29255008753160866,
"grad_norm": 1.171875,
"learning_rate": 0.0004994804446166317,
"loss": 5.8056,
"mean_token_accuracy": 0.139020137488842,
"num_tokens": 6537932.0,
"step": 3760
},
{
"entropy": 5.811407566070557,
"epoch": 0.2929391169033262,
"grad_norm": 1.21875,
"learning_rate": 0.0004994785605093035,
"loss": 5.6878,
"mean_token_accuracy": 0.14702240973711014,
"num_tokens": 6546278.0,
"step": 3765
},
{
"entropy": 5.779412078857422,
"epoch": 0.2933281462750438,
"grad_norm": 1.203125,
"learning_rate": 0.0004994766729958705,
"loss": 5.749,
"mean_token_accuracy": 0.15017975717782975,
"num_tokens": 6554449.0,
"step": 3770
},
{
"entropy": 5.7709815979003904,
"epoch": 0.2937171756467613,
"grad_norm": 1.171875,
"learning_rate": 0.0004994747820763612,
"loss": 5.74,
"mean_token_accuracy": 0.14958301335573196,
"num_tokens": 6563272.0,
"step": 3775
},
{
"entropy": 5.8512287616729735,
"epoch": 0.2941062050184789,
"grad_norm": 1.296875,
"learning_rate": 0.0004994728877508046,
"loss": 5.7479,
"mean_token_accuracy": 0.14908050000667572,
"num_tokens": 6571864.0,
"step": 3780
},
{
"entropy": 5.808426761627198,
"epoch": 0.2944952343901965,
"grad_norm": 1.2578125,
"learning_rate": 0.0004994709900192289,
"loss": 5.7548,
"mean_token_accuracy": 0.15123838260769845,
"num_tokens": 6580435.0,
"step": 3785
},
{
"entropy": 5.829617738723755,
"epoch": 0.294884263761914,
"grad_norm": 1.2734375,
"learning_rate": 0.0004994690888816635,
"loss": 5.7405,
"mean_token_accuracy": 0.15302761197090148,
"num_tokens": 6589261.0,
"step": 3790
},
{
"entropy": 5.815215587615967,
"epoch": 0.2952732931336316,
"grad_norm": 1.2109375,
"learning_rate": 0.0004994671843381367,
"loss": 5.833,
"mean_token_accuracy": 0.14754071235656738,
"num_tokens": 6598057.0,
"step": 3795
},
{
"entropy": 5.89307165145874,
"epoch": 0.29566232250534913,
"grad_norm": 1.15625,
"learning_rate": 0.0004994652763886778,
"loss": 5.7819,
"mean_token_accuracy": 0.14495566934347154,
"num_tokens": 6606439.0,
"step": 3800
},
{
"entropy": 5.779307746887207,
"epoch": 0.2960513518770667,
"grad_norm": 1.234375,
"learning_rate": 0.0004994633650333155,
"loss": 5.7972,
"mean_token_accuracy": 0.150536447763443,
"num_tokens": 6615350.0,
"step": 3805
},
{
"entropy": 5.839250755310059,
"epoch": 0.2964403812487843,
"grad_norm": 1.265625,
"learning_rate": 0.0004994614502720792,
"loss": 5.8374,
"mean_token_accuracy": 0.14464310929179192,
"num_tokens": 6624258.0,
"step": 3810
},
{
"entropy": 5.925694417953491,
"epoch": 0.29682941062050183,
"grad_norm": 1.234375,
"learning_rate": 0.0004994595321049974,
"loss": 5.8396,
"mean_token_accuracy": 0.14261462390422822,
"num_tokens": 6632382.0,
"step": 3815
},
{
"entropy": 5.749386548995972,
"epoch": 0.2972184399922194,
"grad_norm": 1.203125,
"learning_rate": 0.0004994576105320994,
"loss": 5.7459,
"mean_token_accuracy": 0.14289859756827356,
"num_tokens": 6640482.0,
"step": 3820
},
{
"entropy": 5.77762246131897,
"epoch": 0.29760746936393695,
"grad_norm": 1.203125,
"learning_rate": 0.0004994556855534146,
"loss": 5.7249,
"mean_token_accuracy": 0.14987960308790207,
"num_tokens": 6648912.0,
"step": 3825
},
{
"entropy": 5.764063549041748,
"epoch": 0.29799649873565454,
"grad_norm": 1.203125,
"learning_rate": 0.0004994537571689719,
"loss": 5.6934,
"mean_token_accuracy": 0.15349327325820922,
"num_tokens": 6656912.0,
"step": 3830
},
{
"entropy": 5.654044771194458,
"epoch": 0.2983855281073721,
"grad_norm": 1.15625,
"learning_rate": 0.0004994518253788007,
"loss": 5.7261,
"mean_token_accuracy": 0.1497397929430008,
"num_tokens": 6665900.0,
"step": 3835
},
{
"entropy": 5.794566822052002,
"epoch": 0.29877455747908965,
"grad_norm": 1.1796875,
"learning_rate": 0.0004994498901829304,
"loss": 5.7605,
"mean_token_accuracy": 0.1491077497601509,
"num_tokens": 6674426.0,
"step": 3840
},
{
"entropy": 5.859363174438476,
"epoch": 0.29916358685080724,
"grad_norm": 1.1953125,
"learning_rate": 0.0004994479515813903,
"loss": 5.7241,
"mean_token_accuracy": 0.15072896480560302,
"num_tokens": 6683566.0,
"step": 3845
},
{
"entropy": 5.8193401336669925,
"epoch": 0.2995526162225248,
"grad_norm": 1.3828125,
"learning_rate": 0.0004994460095742096,
"loss": 5.7437,
"mean_token_accuracy": 0.14816814959049224,
"num_tokens": 6693341.0,
"step": 3850
},
{
"entropy": 5.744719791412353,
"epoch": 0.29994164559424236,
"grad_norm": 1.2265625,
"learning_rate": 0.000499444064161418,
"loss": 5.6457,
"mean_token_accuracy": 0.14331029802560807,
"num_tokens": 6701734.0,
"step": 3855
},
{
"entropy": 5.705294227600097,
"epoch": 0.30033067496595994,
"grad_norm": 1.1171875,
"learning_rate": 0.000499442115343045,
"loss": 5.6967,
"mean_token_accuracy": 0.14783288091421126,
"num_tokens": 6710572.0,
"step": 3860
},
{
"entropy": 5.857030296325684,
"epoch": 0.3007197043376775,
"grad_norm": 1.1796875,
"learning_rate": 0.0004994401631191202,
"loss": 5.7977,
"mean_token_accuracy": 0.14748454242944717,
"num_tokens": 6719355.0,
"step": 3865
},
{
"entropy": 5.939599275588989,
"epoch": 0.30110873370939506,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994382074896731,
"loss": 5.8436,
"mean_token_accuracy": 0.14460212513804435,
"num_tokens": 6729279.0,
"step": 3870
},
{
"entropy": 5.743366622924805,
"epoch": 0.30149776308111265,
"grad_norm": 1.3125,
"learning_rate": 0.0004994362484547335,
"loss": 5.661,
"mean_token_accuracy": 0.1506835326552391,
"num_tokens": 6737318.0,
"step": 3875
},
{
"entropy": 5.726180648803711,
"epoch": 0.3018867924528302,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994342860143309,
"loss": 5.7294,
"mean_token_accuracy": 0.15141772627830505,
"num_tokens": 6746520.0,
"step": 3880
},
{
"entropy": 5.796295690536499,
"epoch": 0.30227582182454776,
"grad_norm": 1.1875,
"learning_rate": 0.0004994323201684953,
"loss": 5.6717,
"mean_token_accuracy": 0.15030458867549895,
"num_tokens": 6755685.0,
"step": 3885
},
{
"entropy": 5.838547277450561,
"epoch": 0.3026648511962653,
"grad_norm": 1.3046875,
"learning_rate": 0.0004994303509172566,
"loss": 5.8235,
"mean_token_accuracy": 0.1489374205470085,
"num_tokens": 6764347.0,
"step": 3890
},
{
"entropy": 5.891355895996094,
"epoch": 0.3030538805679829,
"grad_norm": 1.359375,
"learning_rate": 0.0004994283782606444,
"loss": 5.7928,
"mean_token_accuracy": 0.14623976349830628,
"num_tokens": 6772887.0,
"step": 3895
},
{
"entropy": 5.719379234313965,
"epoch": 0.30344290993970047,
"grad_norm": 1.2578125,
"learning_rate": 0.0004994264021986888,
"loss": 5.752,
"mean_token_accuracy": 0.14479378163814544,
"num_tokens": 6780860.0,
"step": 3900
},
{
"entropy": 5.743838214874268,
"epoch": 0.303831939311418,
"grad_norm": 1.2421875,
"learning_rate": 0.0004994244227314197,
"loss": 5.7788,
"mean_token_accuracy": 0.14853150248527527,
"num_tokens": 6789491.0,
"step": 3905
},
{
"entropy": 5.849668788909912,
"epoch": 0.3042209686831356,
"grad_norm": 1.3203125,
"learning_rate": 0.0004994224398588672,
"loss": 5.6878,
"mean_token_accuracy": 0.15083892196416854,
"num_tokens": 6797128.0,
"step": 3910
},
{
"entropy": 5.765138626098633,
"epoch": 0.3046099980548531,
"grad_norm": 1.3046875,
"learning_rate": 0.0004994204535810615,
"loss": 5.7392,
"mean_token_accuracy": 0.14893354028463363,
"num_tokens": 6805235.0,
"step": 3915
},
{
"entropy": 5.7339215755462645,
"epoch": 0.3049990274265707,
"grad_norm": 1.234375,
"learning_rate": 0.0004994184638980326,
"loss": 5.6542,
"mean_token_accuracy": 0.14677734673023224,
"num_tokens": 6813863.0,
"step": 3920
},
{
"entropy": 5.850627326965332,
"epoch": 0.3053880567982883,
"grad_norm": 1.1953125,
"learning_rate": 0.0004994164708098107,
"loss": 5.7091,
"mean_token_accuracy": 0.1484633207321167,
"num_tokens": 6822015.0,
"step": 3925
},
{
"entropy": 5.8042628288269045,
"epoch": 0.3057770861700058,
"grad_norm": 1.1953125,
"learning_rate": 0.000499414474316426,
"loss": 5.776,
"mean_token_accuracy": 0.14575532376766204,
"num_tokens": 6830598.0,
"step": 3930
},
{
"entropy": 5.816047525405883,
"epoch": 0.3061661155417234,
"grad_norm": 1.2421875,
"learning_rate": 0.0004994124744179088,
"loss": 5.7568,
"mean_token_accuracy": 0.15651062354445458,
"num_tokens": 6839129.0,
"step": 3935
},
{
"entropy": 5.819494342803955,
"epoch": 0.306555144913441,
"grad_norm": 1.25,
"learning_rate": 0.0004994104711142896,
"loss": 5.7294,
"mean_token_accuracy": 0.14562334567308427,
"num_tokens": 6848199.0,
"step": 3940
},
{
"entropy": 5.744944715499878,
"epoch": 0.3069441742851585,
"grad_norm": 1.21875,
"learning_rate": 0.0004994084644055986,
"loss": 5.7293,
"mean_token_accuracy": 0.1494791552424431,
"num_tokens": 6857340.0,
"step": 3945
},
{
"entropy": 5.801846694946289,
"epoch": 0.3073332036568761,
"grad_norm": 1.2265625,
"learning_rate": 0.0004994064542918664,
"loss": 5.7679,
"mean_token_accuracy": 0.14427911713719369,
"num_tokens": 6865830.0,
"step": 3950
},
{
"entropy": 5.841561651229858,
"epoch": 0.30772223302859364,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994044407731235,
"loss": 5.8153,
"mean_token_accuracy": 0.14207547903060913,
"num_tokens": 6874318.0,
"step": 3955
},
{
"entropy": 5.854608154296875,
"epoch": 0.3081112624003112,
"grad_norm": 1.2578125,
"learning_rate": 0.0004994024238494002,
"loss": 5.8195,
"mean_token_accuracy": 0.1442083813250065,
"num_tokens": 6882008.0,
"step": 3960
},
{
"entropy": 5.7925008773803714,
"epoch": 0.3085002917720288,
"grad_norm": 1.171875,
"learning_rate": 0.0004994004035207274,
"loss": 5.6894,
"mean_token_accuracy": 0.1514485001564026,
"num_tokens": 6890411.0,
"step": 3965
},
{
"entropy": 5.758844566345215,
"epoch": 0.30888932114374634,
"grad_norm": 1.171875,
"learning_rate": 0.0004993983797871356,
"loss": 5.7275,
"mean_token_accuracy": 0.14202559366822243,
"num_tokens": 6899201.0,
"step": 3970
},
{
"entropy": 5.741217756271363,
"epoch": 0.30927835051546393,
"grad_norm": 1.3125,
"learning_rate": 0.0004993963526486555,
"loss": 5.7072,
"mean_token_accuracy": 0.14819234013557434,
"num_tokens": 6907470.0,
"step": 3975
},
{
"entropy": 5.758135032653809,
"epoch": 0.30966737988718146,
"grad_norm": 1.15625,
"learning_rate": 0.000499394322105318,
"loss": 5.7705,
"mean_token_accuracy": 0.14505221620202063,
"num_tokens": 6916542.0,
"step": 3980
},
{
"entropy": 5.811226224899292,
"epoch": 0.31005640925889905,
"grad_norm": 1.1171875,
"learning_rate": 0.0004993922881571538,
"loss": 5.7622,
"mean_token_accuracy": 0.14597265869379045,
"num_tokens": 6925974.0,
"step": 3985
},
{
"entropy": 5.864097452163696,
"epoch": 0.31044543863061663,
"grad_norm": 1.203125,
"learning_rate": 0.0004993902508041939,
"loss": 5.7823,
"mean_token_accuracy": 0.14232797771692277,
"num_tokens": 6934987.0,
"step": 3990
},
{
"entropy": 5.724343395233154,
"epoch": 0.31083446800233416,
"grad_norm": 1.2734375,
"learning_rate": 0.000499388210046469,
"loss": 5.6578,
"mean_token_accuracy": 0.1503308668732643,
"num_tokens": 6944497.0,
"step": 3995
},
{
"entropy": 5.767210245132446,
"epoch": 0.31122349737405175,
"grad_norm": 1.296875,
"learning_rate": 0.0004993861658840102,
"loss": 5.8084,
"mean_token_accuracy": 0.14419659674167634,
"num_tokens": 6953394.0,
"step": 4000
},
{
"entropy": 5.857591390609741,
"epoch": 0.3116125267457693,
"grad_norm": 1.3359375,
"learning_rate": 0.0004993841183168484,
"loss": 5.72,
"mean_token_accuracy": 0.14840124249458314,
"num_tokens": 6961715.0,
"step": 4005
},
{
"entropy": 5.778973197937011,
"epoch": 0.31200155611748687,
"grad_norm": 1.203125,
"learning_rate": 0.0004993820673450148,
"loss": 5.711,
"mean_token_accuracy": 0.15171929597854614,
"num_tokens": 6970041.0,
"step": 4010
},
{
"entropy": 5.807648229598999,
"epoch": 0.31239058548920445,
"grad_norm": 1.1796875,
"learning_rate": 0.0004993800129685404,
"loss": 5.8432,
"mean_token_accuracy": 0.13994109630584717,
"num_tokens": 6979522.0,
"step": 4015
},
{
"entropy": 5.856877374649048,
"epoch": 0.312779614860922,
"grad_norm": 1.2421875,
"learning_rate": 0.0004993779551874565,
"loss": 5.8043,
"mean_token_accuracy": 0.14071302264928817,
"num_tokens": 6988470.0,
"step": 4020
},
{
"entropy": 5.85621976852417,
"epoch": 0.31316864423263957,
"grad_norm": 1.2734375,
"learning_rate": 0.0004993758940017943,
"loss": 5.8328,
"mean_token_accuracy": 0.1494374841451645,
"num_tokens": 6996696.0,
"step": 4025
},
{
"entropy": 5.837846803665161,
"epoch": 0.31355767360435716,
"grad_norm": 1.171875,
"learning_rate": 0.000499373829411585,
"loss": 5.6987,
"mean_token_accuracy": 0.14920781552791595,
"num_tokens": 7005297.0,
"step": 4030
},
{
"entropy": 5.746081399917602,
"epoch": 0.3139467029760747,
"grad_norm": 1.3046875,
"learning_rate": 0.00049937176141686,
"loss": 5.706,
"mean_token_accuracy": 0.14789552241563797,
"num_tokens": 7013569.0,
"step": 4035
},
{
"entropy": 5.712790489196777,
"epoch": 0.3143357323477923,
"grad_norm": 1.2578125,
"learning_rate": 0.0004993696900176506,
"loss": 5.6654,
"mean_token_accuracy": 0.15430878922343255,
"num_tokens": 7021449.0,
"step": 4040
},
{
"entropy": 5.807843160629273,
"epoch": 0.3147247617195098,
"grad_norm": 1.390625,
"learning_rate": 0.0004993676152139883,
"loss": 5.7744,
"mean_token_accuracy": 0.14702282547950746,
"num_tokens": 7029663.0,
"step": 4045
},
{
"entropy": 5.8004063129425045,
"epoch": 0.3151137910912274,
"grad_norm": 1.1953125,
"learning_rate": 0.0004993655370059046,
"loss": 5.7596,
"mean_token_accuracy": 0.14937672913074493,
"num_tokens": 7039038.0,
"step": 4050
},
{
"entropy": 5.813193845748901,
"epoch": 0.315502820462945,
"grad_norm": 1.2109375,
"learning_rate": 0.0004993634553934309,
"loss": 5.7187,
"mean_token_accuracy": 0.14714134261012077,
"num_tokens": 7047595.0,
"step": 4055
},
{
"entropy": 5.806483888626099,
"epoch": 0.3158918498346625,
"grad_norm": 1.25,
"learning_rate": 0.0004993613703765989,
"loss": 5.7476,
"mean_token_accuracy": 0.1446899317204952,
"num_tokens": 7055711.0,
"step": 4060
},
{
"entropy": 5.83050012588501,
"epoch": 0.3162808792063801,
"grad_norm": 1.1484375,
"learning_rate": 0.0004993592819554402,
"loss": 5.7053,
"mean_token_accuracy": 0.15050461292266845,
"num_tokens": 7064100.0,
"step": 4065
},
{
"entropy": 5.730781698226929,
"epoch": 0.3166699085780976,
"grad_norm": 1.21875,
"learning_rate": 0.0004993571901299866,
"loss": 5.6671,
"mean_token_accuracy": 0.15151382014155387,
"num_tokens": 7073040.0,
"step": 4070
},
{
"entropy": 5.802879333496094,
"epoch": 0.3170589379498152,
"grad_norm": 1.40625,
"learning_rate": 0.0004993550949002697,
"loss": 5.7897,
"mean_token_accuracy": 0.14686385542154312,
"num_tokens": 7081513.0,
"step": 4075
},
{
"entropy": 5.71605954170227,
"epoch": 0.3174479673215328,
"grad_norm": 1.28125,
"learning_rate": 0.0004993529962663213,
"loss": 5.6664,
"mean_token_accuracy": 0.1578037440776825,
"num_tokens": 7090704.0,
"step": 4080
},
{
"entropy": 5.719445323944091,
"epoch": 0.31783699669325033,
"grad_norm": 1.109375,
"learning_rate": 0.0004993508942281732,
"loss": 5.8339,
"mean_token_accuracy": 0.14239503517746926,
"num_tokens": 7099692.0,
"step": 4085
},
{
"entropy": 5.829551744461059,
"epoch": 0.3182260260649679,
"grad_norm": 1.28125,
"learning_rate": 0.0004993487887858575,
"loss": 5.6674,
"mean_token_accuracy": 0.14845634177327155,
"num_tokens": 7108021.0,
"step": 4090
},
{
"entropy": 5.787135601043701,
"epoch": 0.31861505543668545,
"grad_norm": 1.3046875,
"learning_rate": 0.000499346679939406,
"loss": 5.7383,
"mean_token_accuracy": 0.15348986834287642,
"num_tokens": 7116689.0,
"step": 4095
},
{
"entropy": 5.830695724487304,
"epoch": 0.31900408480840303,
"grad_norm": 1.375,
"learning_rate": 0.0004993445676888507,
"loss": 5.8181,
"mean_token_accuracy": 0.14692653194069863,
"num_tokens": 7125453.0,
"step": 4100
},
{
"entropy": 5.901071310043335,
"epoch": 0.3193931141801206,
"grad_norm": 1.234375,
"learning_rate": 0.0004993424520342236,
"loss": 5.8614,
"mean_token_accuracy": 0.14383067488670348,
"num_tokens": 7133692.0,
"step": 4105
},
{
"entropy": 5.799852418899536,
"epoch": 0.31978214355183815,
"grad_norm": 1.21875,
"learning_rate": 0.0004993403329755569,
"loss": 5.6846,
"mean_token_accuracy": 0.15570786744356155,
"num_tokens": 7142362.0,
"step": 4110
},
{
"entropy": 5.69540696144104,
"epoch": 0.32017117292355574,
"grad_norm": 1.265625,
"learning_rate": 0.0004993382105128828,
"loss": 5.694,
"mean_token_accuracy": 0.1470162481069565,
"num_tokens": 7151020.0,
"step": 4115
},
{
"entropy": 5.802663040161133,
"epoch": 0.32056020229527327,
"grad_norm": 1.2890625,
"learning_rate": 0.0004993360846462333,
"loss": 5.6156,
"mean_token_accuracy": 0.15607071816921234,
"num_tokens": 7159453.0,
"step": 4120
},
{
"entropy": 5.836108446121216,
"epoch": 0.32094923166699085,
"grad_norm": 1.2578125,
"learning_rate": 0.0004993339553756408,
"loss": 5.8734,
"mean_token_accuracy": 0.14251873642206192,
"num_tokens": 7168758.0,
"step": 4125
},
{
"entropy": 5.747575998306274,
"epoch": 0.32133826103870844,
"grad_norm": 1.328125,
"learning_rate": 0.0004993318227011378,
"loss": 5.6809,
"mean_token_accuracy": 0.14836385846138,
"num_tokens": 7177835.0,
"step": 4130
},
{
"entropy": 5.791856718063355,
"epoch": 0.32172729041042597,
"grad_norm": 1.28125,
"learning_rate": 0.0004993296866227562,
"loss": 5.7434,
"mean_token_accuracy": 0.15285915583372117,
"num_tokens": 7186424.0,
"step": 4135
},
{
"entropy": 5.785421514511109,
"epoch": 0.32211631978214356,
"grad_norm": 1.2109375,
"learning_rate": 0.0004993275471405288,
"loss": 5.7097,
"mean_token_accuracy": 0.15487945079803467,
"num_tokens": 7194800.0,
"step": 4140
},
{
"entropy": 5.753833484649658,
"epoch": 0.32250534915386114,
"grad_norm": 1.203125,
"learning_rate": 0.0004993254042544879,
"loss": 5.624,
"mean_token_accuracy": 0.15332144722342492,
"num_tokens": 7203820.0,
"step": 4145
},
{
"entropy": 5.73719220161438,
"epoch": 0.3228943785255787,
"grad_norm": 1.25,
"learning_rate": 0.000499323257964666,
"loss": 5.6277,
"mean_token_accuracy": 0.1526389017701149,
"num_tokens": 7211793.0,
"step": 4150
},
{
"entropy": 5.769924783706665,
"epoch": 0.32328340789729626,
"grad_norm": 1.171875,
"learning_rate": 0.0004993211082710957,
"loss": 5.6852,
"mean_token_accuracy": 0.15136036276817322,
"num_tokens": 7220471.0,
"step": 4155
},
{
"entropy": 5.725868844985962,
"epoch": 0.3236724372690138,
"grad_norm": 1.265625,
"learning_rate": 0.0004993189551738097,
"loss": 5.7308,
"mean_token_accuracy": 0.14761384800076485,
"num_tokens": 7228788.0,
"step": 4160
},
{
"entropy": 5.8348915576934814,
"epoch": 0.3240614666407314,
"grad_norm": 1.3203125,
"learning_rate": 0.0004993167986728405,
"loss": 5.8991,
"mean_token_accuracy": 0.14883753657341003,
"num_tokens": 7237707.0,
"step": 4165
},
{
"entropy": 5.892117071151733,
"epoch": 0.32445049601244896,
"grad_norm": 1.1875,
"learning_rate": 0.000499314638768221,
"loss": 5.6602,
"mean_token_accuracy": 0.15464966297149657,
"num_tokens": 7246296.0,
"step": 4170
},
{
"entropy": 5.67684326171875,
"epoch": 0.3248395253841665,
"grad_norm": 1.2265625,
"learning_rate": 0.0004993124754599839,
"loss": 5.5873,
"mean_token_accuracy": 0.1556680306792259,
"num_tokens": 7255076.0,
"step": 4175
},
{
"entropy": 5.574007749557495,
"epoch": 0.3252285547558841,
"grad_norm": 1.3515625,
"learning_rate": 0.0004993103087481619,
"loss": 5.6898,
"mean_token_accuracy": 0.14951695650815963,
"num_tokens": 7263169.0,
"step": 4180
},
{
"entropy": 5.827068758010864,
"epoch": 0.3256175841276016,
"grad_norm": 1.171875,
"learning_rate": 0.0004993081386327881,
"loss": 5.7499,
"mean_token_accuracy": 0.15402725636959075,
"num_tokens": 7271748.0,
"step": 4185
},
{
"entropy": 5.725684070587159,
"epoch": 0.3260066134993192,
"grad_norm": 1.1875,
"learning_rate": 0.0004993059651138953,
"loss": 5.5602,
"mean_token_accuracy": 0.1556816130876541,
"num_tokens": 7280291.0,
"step": 4190
},
{
"entropy": 5.69296875,
"epoch": 0.3263956428710368,
"grad_norm": 1.2265625,
"learning_rate": 0.0004993037881915165,
"loss": 5.7556,
"mean_token_accuracy": 0.14534951150417327,
"num_tokens": 7290307.0,
"step": 4195
},
{
"entropy": 5.911472511291504,
"epoch": 0.3267846722427543,
"grad_norm": 1.2734375,
"learning_rate": 0.0004993016078656847,
"loss": 5.8196,
"mean_token_accuracy": 0.14084672108292579,
"num_tokens": 7297942.0,
"step": 4200
},
{
"entropy": 5.713421821594238,
"epoch": 0.3271737016144719,
"grad_norm": 1.2109375,
"learning_rate": 0.000499299424136433,
"loss": 5.703,
"mean_token_accuracy": 0.14439374804496766,
"num_tokens": 7307378.0,
"step": 4205
},
{
"entropy": 5.755586624145508,
"epoch": 0.32756273098618943,
"grad_norm": 1.2890625,
"learning_rate": 0.0004992972370037946,
"loss": 5.6635,
"mean_token_accuracy": 0.1554426446557045,
"num_tokens": 7315985.0,
"step": 4210
},
{
"entropy": 5.809132528305054,
"epoch": 0.327951760357907,
"grad_norm": 1.2578125,
"learning_rate": 0.0004992950464678026,
"loss": 5.7924,
"mean_token_accuracy": 0.14466845989227295,
"num_tokens": 7324142.0,
"step": 4215
},
{
"entropy": 5.739095306396484,
"epoch": 0.3283407897296246,
"grad_norm": 1.234375,
"learning_rate": 0.0004992928525284903,
"loss": 5.6415,
"mean_token_accuracy": 0.14974627569317817,
"num_tokens": 7333012.0,
"step": 4220
},
{
"entropy": 5.764504671096802,
"epoch": 0.32872981910134214,
"grad_norm": 1.2421875,
"learning_rate": 0.000499290655185891,
"loss": 5.6733,
"mean_token_accuracy": 0.1509619578719139,
"num_tokens": 7341838.0,
"step": 4225
},
{
"entropy": 5.677153444290161,
"epoch": 0.3291188484730597,
"grad_norm": 1.25,
"learning_rate": 0.000499288454440038,
"loss": 5.6118,
"mean_token_accuracy": 0.16084043234586715,
"num_tokens": 7349986.0,
"step": 4230
},
{
"entropy": 5.832826709747314,
"epoch": 0.3295078778447773,
"grad_norm": 1.234375,
"learning_rate": 0.0004992862502909645,
"loss": 5.8509,
"mean_token_accuracy": 0.14144516214728356,
"num_tokens": 7359428.0,
"step": 4235
},
{
"entropy": 5.873200845718384,
"epoch": 0.32989690721649484,
"grad_norm": 1.21875,
"learning_rate": 0.0004992840427387043,
"loss": 5.7791,
"mean_token_accuracy": 0.14769574701786042,
"num_tokens": 7368524.0,
"step": 4240
},
{
"entropy": 5.753816413879394,
"epoch": 0.3302859365882124,
"grad_norm": 1.09375,
"learning_rate": 0.0004992818317832908,
"loss": 5.6531,
"mean_token_accuracy": 0.1480708360671997,
"num_tokens": 7377979.0,
"step": 4245
},
{
"entropy": 5.716938257217407,
"epoch": 0.33067496595992996,
"grad_norm": 1.2421875,
"learning_rate": 0.0004992796174247574,
"loss": 5.5792,
"mean_token_accuracy": 0.15270080417394638,
"num_tokens": 7386283.0,
"step": 4250
},
{
"entropy": 5.703174495697022,
"epoch": 0.33106399533164754,
"grad_norm": 1.203125,
"learning_rate": 0.0004992773996631378,
"loss": 5.7083,
"mean_token_accuracy": 0.14178377091884614,
"num_tokens": 7395409.0,
"step": 4255
},
{
"entropy": 5.793183040618897,
"epoch": 0.33145302470336513,
"grad_norm": 1.2734375,
"learning_rate": 0.0004992751784984656,
"loss": 5.7763,
"mean_token_accuracy": 0.14899417757987976,
"num_tokens": 7403389.0,
"step": 4260
},
{
"entropy": 5.7782896041870115,
"epoch": 0.33184205407508266,
"grad_norm": 1.2109375,
"learning_rate": 0.0004992729539307745,
"loss": 5.703,
"mean_token_accuracy": 0.15201039761304855,
"num_tokens": 7411506.0,
"step": 4265
},
{
"entropy": 5.812991285324097,
"epoch": 0.33223108344680025,
"grad_norm": 1.25,
"learning_rate": 0.0004992707259600984,
"loss": 5.7846,
"mean_token_accuracy": 0.14998362809419633,
"num_tokens": 7420219.0,
"step": 4270
},
{
"entropy": 5.757145166397095,
"epoch": 0.3326201128185178,
"grad_norm": 1.25,
"learning_rate": 0.0004992684945864709,
"loss": 5.7129,
"mean_token_accuracy": 0.15293968468904495,
"num_tokens": 7428442.0,
"step": 4275
},
{
"entropy": 5.798875713348389,
"epoch": 0.33300914219023536,
"grad_norm": 1.2109375,
"learning_rate": 0.0004992662598099261,
"loss": 5.6653,
"mean_token_accuracy": 0.1510924369096756,
"num_tokens": 7436364.0,
"step": 4280
},
{
"entropy": 5.772172117233277,
"epoch": 0.33339817156195295,
"grad_norm": 1.2578125,
"learning_rate": 0.0004992640216304975,
"loss": 5.7325,
"mean_token_accuracy": 0.14351431503891945,
"num_tokens": 7444122.0,
"step": 4285
},
{
"entropy": 5.835605764389038,
"epoch": 0.3337872009336705,
"grad_norm": 1.25,
"learning_rate": 0.0004992617800482196,
"loss": 5.7413,
"mean_token_accuracy": 0.14694980010390282,
"num_tokens": 7454122.0,
"step": 4290
},
{
"entropy": 5.806734895706176,
"epoch": 0.33417623030538807,
"grad_norm": 1.328125,
"learning_rate": 0.000499259535063126,
"loss": 5.752,
"mean_token_accuracy": 0.15015026181936264,
"num_tokens": 7463057.0,
"step": 4295
},
{
"entropy": 5.819530391693116,
"epoch": 0.3345652596771056,
"grad_norm": 1.1796875,
"learning_rate": 0.000499257286675251,
"loss": 5.7053,
"mean_token_accuracy": 0.14335729479789733,
"num_tokens": 7471675.0,
"step": 4300
},
{
"entropy": 5.699312400817871,
"epoch": 0.3349542890488232,
"grad_norm": 1.1953125,
"learning_rate": 0.0004992550348846285,
"loss": 5.6546,
"mean_token_accuracy": 0.1525387153029442,
"num_tokens": 7479870.0,
"step": 4305
},
{
"entropy": 5.7911989212036135,
"epoch": 0.33534331842054077,
"grad_norm": 1.1953125,
"learning_rate": 0.0004992527796912928,
"loss": 5.7359,
"mean_token_accuracy": 0.14847930446267127,
"num_tokens": 7489830.0,
"step": 4310
},
{
"entropy": 5.775819587707519,
"epoch": 0.3357323477922583,
"grad_norm": 1.265625,
"learning_rate": 0.0004992505210952782,
"loss": 5.6581,
"mean_token_accuracy": 0.1594860151410103,
"num_tokens": 7497852.0,
"step": 4315
},
{
"entropy": 5.789813709259033,
"epoch": 0.3361213771639759,
"grad_norm": 1.2734375,
"learning_rate": 0.0004992482590966188,
"loss": 5.8037,
"mean_token_accuracy": 0.14984895661473274,
"num_tokens": 7506419.0,
"step": 4320
},
{
"entropy": 5.7952008724212645,
"epoch": 0.3365104065356934,
"grad_norm": 1.1796875,
"learning_rate": 0.0004992459936953489,
"loss": 5.6975,
"mean_token_accuracy": 0.15465806424617767,
"num_tokens": 7514757.0,
"step": 4325
},
{
"entropy": 5.809590005874634,
"epoch": 0.336899435907411,
"grad_norm": 1.2265625,
"learning_rate": 0.0004992437248915032,
"loss": 5.7543,
"mean_token_accuracy": 0.14698707759380342,
"num_tokens": 7523912.0,
"step": 4330
},
{
"entropy": 5.772178649902344,
"epoch": 0.3372884652791286,
"grad_norm": 1.1875,
"learning_rate": 0.0004992414526851157,
"loss": 5.6248,
"mean_token_accuracy": 0.15759740620851517,
"num_tokens": 7532401.0,
"step": 4335
},
{
"entropy": 5.6291059970855715,
"epoch": 0.3376774946508461,
"grad_norm": 1.421875,
"learning_rate": 0.0004992391770762212,
"loss": 5.599,
"mean_token_accuracy": 0.15522923022508622,
"num_tokens": 7541257.0,
"step": 4340
},
{
"entropy": 5.672086715698242,
"epoch": 0.3380665240225637,
"grad_norm": 1.3046875,
"learning_rate": 0.000499236898064854,
"loss": 5.5396,
"mean_token_accuracy": 0.1525227814912796,
"num_tokens": 7549812.0,
"step": 4345
},
{
"entropy": 5.688179683685303,
"epoch": 0.3384555533942813,
"grad_norm": 1.2265625,
"learning_rate": 0.0004992346156510488,
"loss": 5.6406,
"mean_token_accuracy": 0.15223049223423005,
"num_tokens": 7559319.0,
"step": 4350
},
{
"entropy": 5.756983423233033,
"epoch": 0.3388445827659988,
"grad_norm": 1.1796875,
"learning_rate": 0.0004992323298348403,
"loss": 5.7759,
"mean_token_accuracy": 0.14724590629339218,
"num_tokens": 7567900.0,
"step": 4355
},
{
"entropy": 5.881202983856201,
"epoch": 0.3392336121377164,
"grad_norm": 1.2421875,
"learning_rate": 0.000499230040616263,
"loss": 5.7154,
"mean_token_accuracy": 0.1452577978372574,
"num_tokens": 7576163.0,
"step": 4360
},
{
"entropy": 5.70105996131897,
"epoch": 0.33962264150943394,
"grad_norm": 1.234375,
"learning_rate": 0.0004992277479953518,
"loss": 5.6286,
"mean_token_accuracy": 0.15342977792024612,
"num_tokens": 7584221.0,
"step": 4365
},
{
"entropy": 5.741499710083008,
"epoch": 0.34001167088115153,
"grad_norm": 1.2734375,
"learning_rate": 0.0004992254519721414,
"loss": 5.6279,
"mean_token_accuracy": 0.1542896419763565,
"num_tokens": 7593569.0,
"step": 4370
},
{
"entropy": 5.776559448242187,
"epoch": 0.3404007002528691,
"grad_norm": 1.2421875,
"learning_rate": 0.0004992231525466666,
"loss": 5.6975,
"mean_token_accuracy": 0.15335093662142754,
"num_tokens": 7602666.0,
"step": 4375
},
{
"entropy": 5.710270786285401,
"epoch": 0.34078972962458665,
"grad_norm": 1.2890625,
"learning_rate": 0.0004992208497189624,
"loss": 5.7656,
"mean_token_accuracy": 0.14594510644674302,
"num_tokens": 7611874.0,
"step": 4380
},
{
"entropy": 5.758926343917847,
"epoch": 0.34117875899630423,
"grad_norm": 1.265625,
"learning_rate": 0.0004992185434890637,
"loss": 5.7198,
"mean_token_accuracy": 0.15153981149196624,
"num_tokens": 7620385.0,
"step": 4385
},
{
"entropy": 5.837249088287353,
"epoch": 0.34156778836802176,
"grad_norm": 1.2421875,
"learning_rate": 0.0004992162338570055,
"loss": 5.7949,
"mean_token_accuracy": 0.13906663209199904,
"num_tokens": 7628786.0,
"step": 4390
},
{
"entropy": 5.7165384769439695,
"epoch": 0.34195681773973935,
"grad_norm": 1.1796875,
"learning_rate": 0.0004992139208228227,
"loss": 5.5799,
"mean_token_accuracy": 0.1546400859951973,
"num_tokens": 7637389.0,
"step": 4395
},
{
"entropy": 5.71855788230896,
"epoch": 0.34234584711145694,
"grad_norm": 1.515625,
"learning_rate": 0.0004992116043865506,
"loss": 5.7634,
"mean_token_accuracy": 0.14998725801706314,
"num_tokens": 7645793.0,
"step": 4400
},
{
"entropy": 5.859033393859863,
"epoch": 0.34273487648317447,
"grad_norm": 1.2890625,
"learning_rate": 0.0004992092845482244,
"loss": 5.6981,
"mean_token_accuracy": 0.15337317287921906,
"num_tokens": 7653951.0,
"step": 4405
},
{
"entropy": 5.665503597259521,
"epoch": 0.34312390585489205,
"grad_norm": 1.1640625,
"learning_rate": 0.0004992069613078791,
"loss": 5.5882,
"mean_token_accuracy": 0.1616547703742981,
"num_tokens": 7662753.0,
"step": 4410
},
{
"entropy": 5.646042823791504,
"epoch": 0.3435129352266096,
"grad_norm": 1.3046875,
"learning_rate": 0.0004992046346655499,
"loss": 5.5503,
"mean_token_accuracy": 0.15094086453318595,
"num_tokens": 7671832.0,
"step": 4415
},
{
"entropy": 5.803126287460327,
"epoch": 0.34390196459832717,
"grad_norm": 1.21875,
"learning_rate": 0.0004992023046212723,
"loss": 5.7209,
"mean_token_accuracy": 0.14602495953440667,
"num_tokens": 7680969.0,
"step": 4420
},
{
"entropy": 5.8474115371704105,
"epoch": 0.34429099397004476,
"grad_norm": 1.28125,
"learning_rate": 0.0004991999711750816,
"loss": 5.7295,
"mean_token_accuracy": 0.14345005601644517,
"num_tokens": 7689277.0,
"step": 4425
},
{
"entropy": 5.678358888626098,
"epoch": 0.3446800233417623,
"grad_norm": 1.296875,
"learning_rate": 0.0004991976343270133,
"loss": 5.6502,
"mean_token_accuracy": 0.15776083916425704,
"num_tokens": 7697373.0,
"step": 4430
},
{
"entropy": 5.839732694625854,
"epoch": 0.3450690527134799,
"grad_norm": 1.296875,
"learning_rate": 0.0004991952940771026,
"loss": 5.73,
"mean_token_accuracy": 0.1477728880941868,
"num_tokens": 7705767.0,
"step": 4435
},
{
"entropy": 5.745546770095825,
"epoch": 0.34545808208519746,
"grad_norm": 1.34375,
"learning_rate": 0.0004991929504253852,
"loss": 5.5992,
"mean_token_accuracy": 0.1563803017139435,
"num_tokens": 7713746.0,
"step": 4440
},
{
"entropy": 5.654825639724732,
"epoch": 0.345847111456915,
"grad_norm": 1.328125,
"learning_rate": 0.0004991906033718966,
"loss": 5.7396,
"mean_token_accuracy": 0.14519762098789216,
"num_tokens": 7721829.0,
"step": 4445
},
{
"entropy": 5.703803396224975,
"epoch": 0.3462361408286326,
"grad_norm": 1.21875,
"learning_rate": 0.0004991882529166724,
"loss": 5.6408,
"mean_token_accuracy": 0.15357706546783448,
"num_tokens": 7730479.0,
"step": 4450
},
{
"entropy": 5.735348558425903,
"epoch": 0.3466251702003501,
"grad_norm": 1.21875,
"learning_rate": 0.0004991858990597483,
"loss": 5.5983,
"mean_token_accuracy": 0.1547751769423485,
"num_tokens": 7738659.0,
"step": 4455
},
{
"entropy": 5.813489246368408,
"epoch": 0.3470141995720677,
"grad_norm": 1.2890625,
"learning_rate": 0.0004991835418011601,
"loss": 5.8591,
"mean_token_accuracy": 0.14898139089345933,
"num_tokens": 7747597.0,
"step": 4460
},
{
"entropy": 5.801498508453369,
"epoch": 0.3474032289437853,
"grad_norm": 1.25,
"learning_rate": 0.0004991811811409434,
"loss": 5.7076,
"mean_token_accuracy": 0.14893003180623055,
"num_tokens": 7756666.0,
"step": 4465
},
{
"entropy": 5.805731916427613,
"epoch": 0.3477922583155028,
"grad_norm": 1.2109375,
"learning_rate": 0.0004991788170791341,
"loss": 5.7394,
"mean_token_accuracy": 0.14709567725658418,
"num_tokens": 7765621.0,
"step": 4470
},
{
"entropy": 5.706645917892456,
"epoch": 0.3481812876872204,
"grad_norm": 1.234375,
"learning_rate": 0.0004991764496157682,
"loss": 5.633,
"mean_token_accuracy": 0.149008846282959,
"num_tokens": 7774137.0,
"step": 4475
},
{
"entropy": 5.792159605026245,
"epoch": 0.34857031705893793,
"grad_norm": 1.2265625,
"learning_rate": 0.0004991740787508814,
"loss": 5.7491,
"mean_token_accuracy": 0.14444359689950942,
"num_tokens": 7783144.0,
"step": 4480
},
{
"entropy": 5.792681503295898,
"epoch": 0.3489593464306555,
"grad_norm": 1.28125,
"learning_rate": 0.0004991717044845097,
"loss": 5.7591,
"mean_token_accuracy": 0.14241152629256248,
"num_tokens": 7791530.0,
"step": 4485
},
{
"entropy": 5.733056831359863,
"epoch": 0.3493483758023731,
"grad_norm": 1.2265625,
"learning_rate": 0.0004991693268166892,
"loss": 5.6294,
"mean_token_accuracy": 0.15588642954826354,
"num_tokens": 7800869.0,
"step": 4490
},
{
"entropy": 5.743151617050171,
"epoch": 0.34973740517409063,
"grad_norm": 1.2578125,
"learning_rate": 0.0004991669457474559,
"loss": 5.702,
"mean_token_accuracy": 0.15250983089208603,
"num_tokens": 7810201.0,
"step": 4495
},
{
"entropy": 5.70795955657959,
"epoch": 0.3501264345458082,
"grad_norm": 1.1796875,
"learning_rate": 0.0004991645612768461,
"loss": 5.6051,
"mean_token_accuracy": 0.15941296368837357,
"num_tokens": 7818529.0,
"step": 4500
},
{
"entropy": 5.727574253082276,
"epoch": 0.35051546391752575,
"grad_norm": 1.2421875,
"learning_rate": 0.0004991621734048958,
"loss": 5.6513,
"mean_token_accuracy": 0.15623073428869247,
"num_tokens": 7827254.0,
"step": 4505
},
{
"entropy": 5.767280435562133,
"epoch": 0.35090449328924334,
"grad_norm": 1.2109375,
"learning_rate": 0.0004991597821316415,
"loss": 5.6719,
"mean_token_accuracy": 0.15514668226242065,
"num_tokens": 7836437.0,
"step": 4510
},
{
"entropy": 5.724820709228515,
"epoch": 0.3512935226609609,
"grad_norm": 1.1875,
"learning_rate": 0.0004991573874571192,
"loss": 5.6718,
"mean_token_accuracy": 0.1533803641796112,
"num_tokens": 7844939.0,
"step": 4515
},
{
"entropy": 5.74892292022705,
"epoch": 0.35168255203267845,
"grad_norm": 1.25,
"learning_rate": 0.0004991549893813653,
"loss": 5.6687,
"mean_token_accuracy": 0.15008964389562607,
"num_tokens": 7852889.0,
"step": 4520
},
{
"entropy": 5.7785228252410885,
"epoch": 0.35207158140439604,
"grad_norm": 1.28125,
"learning_rate": 0.0004991525879044162,
"loss": 5.7899,
"mean_token_accuracy": 0.14683696180582045,
"num_tokens": 7861867.0,
"step": 4525
},
{
"entropy": 5.77210807800293,
"epoch": 0.35246061077611357,
"grad_norm": 1.328125,
"learning_rate": 0.0004991501830263084,
"loss": 5.6071,
"mean_token_accuracy": 0.15673087388277054,
"num_tokens": 7869343.0,
"step": 4530
},
{
"entropy": 5.753970432281494,
"epoch": 0.35284964014783116,
"grad_norm": 1.3125,
"learning_rate": 0.0004991477747470783,
"loss": 5.636,
"mean_token_accuracy": 0.15080263912677766,
"num_tokens": 7877142.0,
"step": 4535
},
{
"entropy": 5.708729314804077,
"epoch": 0.35323866951954874,
"grad_norm": 1.375,
"learning_rate": 0.0004991453630667625,
"loss": 5.6686,
"mean_token_accuracy": 0.15797902941703795,
"num_tokens": 7885183.0,
"step": 4540
},
{
"entropy": 5.585329294204712,
"epoch": 0.3536276988912663,
"grad_norm": 1.3828125,
"learning_rate": 0.0004991429479853976,
"loss": 5.4989,
"mean_token_accuracy": 0.16130374521017074,
"num_tokens": 7893848.0,
"step": 4545
},
{
"entropy": 5.697960329055786,
"epoch": 0.35401672826298386,
"grad_norm": 1.2578125,
"learning_rate": 0.0004991405295030202,
"loss": 5.6698,
"mean_token_accuracy": 0.15179314315319062,
"num_tokens": 7902438.0,
"step": 4550
},
{
"entropy": 5.731406307220459,
"epoch": 0.35440575763470145,
"grad_norm": 1.4375,
"learning_rate": 0.0004991381076196671,
"loss": 5.6484,
"mean_token_accuracy": 0.15274685472249985,
"num_tokens": 7910951.0,
"step": 4555
},
{
"entropy": 5.7294269561767575,
"epoch": 0.354794787006419,
"grad_norm": 1.328125,
"learning_rate": 0.0004991356823353748,
"loss": 5.7549,
"mean_token_accuracy": 0.1446766048669815,
"num_tokens": 7920375.0,
"step": 4560
},
{
"entropy": 5.696258449554444,
"epoch": 0.35518381637813656,
"grad_norm": 1.3203125,
"learning_rate": 0.0004991332536501804,
"loss": 5.6084,
"mean_token_accuracy": 0.159688700735569,
"num_tokens": 7928656.0,
"step": 4565
},
{
"entropy": 5.8622143268585205,
"epoch": 0.3555728457498541,
"grad_norm": 1.296875,
"learning_rate": 0.0004991308215641205,
"loss": 5.8611,
"mean_token_accuracy": 0.14252166226506233,
"num_tokens": 7936929.0,
"step": 4570
},
{
"entropy": 5.791495895385742,
"epoch": 0.3559618751215717,
"grad_norm": 1.2421875,
"learning_rate": 0.0004991283860772323,
"loss": 5.6697,
"mean_token_accuracy": 0.14830510020256044,
"num_tokens": 7945592.0,
"step": 4575
},
{
"entropy": 5.7066380977630615,
"epoch": 0.35635090449328927,
"grad_norm": 1.34375,
"learning_rate": 0.0004991259471895525,
"loss": 5.6207,
"mean_token_accuracy": 0.16197174191474914,
"num_tokens": 7953656.0,
"step": 4580
},
{
"entropy": 5.781188344955444,
"epoch": 0.3567399338650068,
"grad_norm": 1.3046875,
"learning_rate": 0.0004991235049011182,
"loss": 5.6767,
"mean_token_accuracy": 0.15540962666273117,
"num_tokens": 7962038.0,
"step": 4585
},
{
"entropy": 5.787206315994263,
"epoch": 0.3571289632367244,
"grad_norm": 1.28125,
"learning_rate": 0.0004991210592119663,
"loss": 5.7694,
"mean_token_accuracy": 0.14897529482841493,
"num_tokens": 7970338.0,
"step": 4590
},
{
"entropy": 5.7750184535980225,
"epoch": 0.3575179926084419,
"grad_norm": 1.3125,
"learning_rate": 0.0004991186101221342,
"loss": 5.6861,
"mean_token_accuracy": 0.15258906185626983,
"num_tokens": 7979173.0,
"step": 4595
},
{
"entropy": 5.662876081466675,
"epoch": 0.3579070219801595,
"grad_norm": 1.4609375,
"learning_rate": 0.000499116157631659,
"loss": 5.5678,
"mean_token_accuracy": 0.1568956859409809,
"num_tokens": 7987245.0,
"step": 4600
},
{
"entropy": 5.838749551773072,
"epoch": 0.3582960513518771,
"grad_norm": 1.4453125,
"learning_rate": 0.0004991137017405777,
"loss": 5.7822,
"mean_token_accuracy": 0.1373494178056717,
"num_tokens": 7995551.0,
"step": 4605
},
{
"entropy": 5.741954660415649,
"epoch": 0.3586850807235946,
"grad_norm": 1.265625,
"learning_rate": 0.0004991112424489277,
"loss": 5.5894,
"mean_token_accuracy": 0.1502602607011795,
"num_tokens": 8004069.0,
"step": 4610
},
{
"entropy": 5.668931388854981,
"epoch": 0.3590741100953122,
"grad_norm": 1.3046875,
"learning_rate": 0.0004991087797567462,
"loss": 5.6494,
"mean_token_accuracy": 0.15510962158441544,
"num_tokens": 8012537.0,
"step": 4615
},
{
"entropy": 5.768225383758545,
"epoch": 0.35946313946702974,
"grad_norm": 1.25,
"learning_rate": 0.0004991063136640709,
"loss": 5.6506,
"mean_token_accuracy": 0.15651883333921432,
"num_tokens": 8021387.0,
"step": 4620
},
{
"entropy": 5.744030857086182,
"epoch": 0.3598521688387473,
"grad_norm": 1.34375,
"learning_rate": 0.0004991038441709388,
"loss": 5.6207,
"mean_token_accuracy": 0.15271022394299508,
"num_tokens": 8030261.0,
"step": 4625
},
{
"entropy": 5.757571983337402,
"epoch": 0.3602411982104649,
"grad_norm": 1.265625,
"learning_rate": 0.0004991013712773877,
"loss": 5.6527,
"mean_token_accuracy": 0.15131479799747466,
"num_tokens": 8038161.0,
"step": 4630
},
{
"entropy": 5.704322004318238,
"epoch": 0.36063022758218244,
"grad_norm": 1.3359375,
"learning_rate": 0.0004990988949834549,
"loss": 5.7469,
"mean_token_accuracy": 0.15158456489443778,
"num_tokens": 8046340.0,
"step": 4635
},
{
"entropy": 5.742067337036133,
"epoch": 0.3610192569539,
"grad_norm": 1.3203125,
"learning_rate": 0.000499096415289178,
"loss": 5.6773,
"mean_token_accuracy": 0.15533811300992967,
"num_tokens": 8054997.0,
"step": 4640
},
{
"entropy": 5.707707309722901,
"epoch": 0.3614082863256176,
"grad_norm": 1.171875,
"learning_rate": 0.0004990939321945948,
"loss": 5.6061,
"mean_token_accuracy": 0.16247581839561462,
"num_tokens": 8063686.0,
"step": 4645
},
{
"entropy": 5.7515209197998045,
"epoch": 0.36179731569733514,
"grad_norm": 1.234375,
"learning_rate": 0.0004990914456997427,
"loss": 5.7188,
"mean_token_accuracy": 0.14975331276655196,
"num_tokens": 8072999.0,
"step": 4650
},
{
"entropy": 5.686860656738281,
"epoch": 0.36218634506905273,
"grad_norm": 1.28125,
"learning_rate": 0.0004990889558046598,
"loss": 5.5856,
"mean_token_accuracy": 0.1535652443766594,
"num_tokens": 8081596.0,
"step": 4655
},
{
"entropy": 5.735697031021118,
"epoch": 0.36257537444077026,
"grad_norm": 1.2578125,
"learning_rate": 0.0004990864625093836,
"loss": 5.8266,
"mean_token_accuracy": 0.14313812404870987,
"num_tokens": 8090948.0,
"step": 4660
},
{
"entropy": 5.772059965133667,
"epoch": 0.36296440381248785,
"grad_norm": 1.3046875,
"learning_rate": 0.0004990839658139519,
"loss": 5.8067,
"mean_token_accuracy": 0.14405996352434158,
"num_tokens": 8099445.0,
"step": 4665
},
{
"entropy": 5.783945560455322,
"epoch": 0.36335343318420543,
"grad_norm": 1.2265625,
"learning_rate": 0.0004990814657184028,
"loss": 5.6526,
"mean_token_accuracy": 0.14939153492450713,
"num_tokens": 8108119.0,
"step": 4670
},
{
"entropy": 5.717726993560791,
"epoch": 0.36374246255592296,
"grad_norm": 1.203125,
"learning_rate": 0.0004990789622227741,
"loss": 5.6631,
"mean_token_accuracy": 0.15249117463827133,
"num_tokens": 8117060.0,
"step": 4675
},
{
"entropy": 5.716911125183105,
"epoch": 0.36413149192764055,
"grad_norm": 1.6875,
"learning_rate": 0.0004990764553271038,
"loss": 5.6576,
"mean_token_accuracy": 0.15026922672986984,
"num_tokens": 8126229.0,
"step": 4680
},
{
"entropy": 5.766642808914185,
"epoch": 0.3645205212993581,
"grad_norm": 1.2890625,
"learning_rate": 0.0004990739450314299,
"loss": 5.6014,
"mean_token_accuracy": 0.16554867178201677,
"num_tokens": 8134694.0,
"step": 4685
},
{
"entropy": 5.680392742156982,
"epoch": 0.36490955067107567,
"grad_norm": 1.2421875,
"learning_rate": 0.0004990714313357906,
"loss": 5.6781,
"mean_token_accuracy": 0.1550322487950325,
"num_tokens": 8143329.0,
"step": 4690
},
{
"entropy": 5.740098524093628,
"epoch": 0.36529858004279325,
"grad_norm": 1.2578125,
"learning_rate": 0.000499068914240224,
"loss": 5.6661,
"mean_token_accuracy": 0.1539440505206585,
"num_tokens": 8151432.0,
"step": 4695
},
{
"entropy": 5.751394510269165,
"epoch": 0.3656876094145108,
"grad_norm": 1.28125,
"learning_rate": 0.0004990663937447682,
"loss": 5.6257,
"mean_token_accuracy": 0.14900294989347457,
"num_tokens": 8159677.0,
"step": 4700
},
{
"entropy": 5.672762489318847,
"epoch": 0.36607663878622837,
"grad_norm": 1.265625,
"learning_rate": 0.0004990638698494615,
"loss": 5.7057,
"mean_token_accuracy": 0.15114860534667968,
"num_tokens": 8168710.0,
"step": 4705
},
{
"entropy": 5.734380531311035,
"epoch": 0.3664656681579459,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990613425543423,
"loss": 5.6783,
"mean_token_accuracy": 0.15022775977849961,
"num_tokens": 8177501.0,
"step": 4710
},
{
"entropy": 5.835424184799194,
"epoch": 0.3668546975296635,
"grad_norm": 1.296875,
"learning_rate": 0.0004990588118594487,
"loss": 5.7708,
"mean_token_accuracy": 0.14951381012797355,
"num_tokens": 8185791.0,
"step": 4715
},
{
"entropy": 5.801257944107055,
"epoch": 0.3672437269013811,
"grad_norm": 1.3671875,
"learning_rate": 0.0004990562777648194,
"loss": 5.7882,
"mean_token_accuracy": 0.14626498967409135,
"num_tokens": 8195020.0,
"step": 4720
},
{
"entropy": 5.712733125686645,
"epoch": 0.3676327562730986,
"grad_norm": 1.28125,
"learning_rate": 0.0004990537402704928,
"loss": 5.652,
"mean_token_accuracy": 0.1529953584074974,
"num_tokens": 8203540.0,
"step": 4725
},
{
"entropy": 5.784476709365845,
"epoch": 0.3680217856448162,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990511993765071,
"loss": 5.6471,
"mean_token_accuracy": 0.14924908876419068,
"num_tokens": 8211985.0,
"step": 4730
},
{
"entropy": 5.748035717010498,
"epoch": 0.3684108150165337,
"grad_norm": 1.25,
"learning_rate": 0.0004990486550829011,
"loss": 5.7379,
"mean_token_accuracy": 0.15286693423986436,
"num_tokens": 8220289.0,
"step": 4735
},
{
"entropy": 5.825424718856811,
"epoch": 0.3687998443882513,
"grad_norm": 1.2109375,
"learning_rate": 0.0004990461073897134,
"loss": 5.7497,
"mean_token_accuracy": 0.14440721794962882,
"num_tokens": 8229652.0,
"step": 4740
},
{
"entropy": 5.832679271697998,
"epoch": 0.3691888737599689,
"grad_norm": 1.28125,
"learning_rate": 0.0004990435562969827,
"loss": 5.7509,
"mean_token_accuracy": 0.14374158829450606,
"num_tokens": 8237728.0,
"step": 4745
},
{
"entropy": 5.732454919815064,
"epoch": 0.3695779031316864,
"grad_norm": 1.2734375,
"learning_rate": 0.0004990410018047475,
"loss": 5.6361,
"mean_token_accuracy": 0.15370055437088012,
"num_tokens": 8246393.0,
"step": 4750
},
{
"entropy": 5.723031377792358,
"epoch": 0.369966932503404,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990384439130467,
"loss": 5.6098,
"mean_token_accuracy": 0.1546899065375328,
"num_tokens": 8256736.0,
"step": 4755
},
{
"entropy": 5.770855712890625,
"epoch": 0.3703559618751216,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990358826219192,
"loss": 5.6885,
"mean_token_accuracy": 0.15330916792154312,
"num_tokens": 8265126.0,
"step": 4760
},
{
"entropy": 5.6247711181640625,
"epoch": 0.37074499124683913,
"grad_norm": 1.2890625,
"learning_rate": 0.0004990333179314038,
"loss": 5.5857,
"mean_token_accuracy": 0.1569363385438919,
"num_tokens": 8273136.0,
"step": 4765
},
{
"entropy": 5.641842555999756,
"epoch": 0.3711340206185567,
"grad_norm": 1.265625,
"learning_rate": 0.0004990307498415393,
"loss": 5.5334,
"mean_token_accuracy": 0.15607051327824592,
"num_tokens": 8281697.0,
"step": 4770
},
{
"entropy": 5.676371383666992,
"epoch": 0.37152304999027425,
"grad_norm": 1.2734375,
"learning_rate": 0.0004990281783523648,
"loss": 5.6734,
"mean_token_accuracy": 0.15302693992853164,
"num_tokens": 8290765.0,
"step": 4775
},
{
"entropy": 5.817877292633057,
"epoch": 0.37191207936199183,
"grad_norm": 1.2734375,
"learning_rate": 0.0004990256034639192,
"loss": 5.7627,
"mean_token_accuracy": 0.14589882269501686,
"num_tokens": 8299875.0,
"step": 4780
},
{
"entropy": 5.781397914886474,
"epoch": 0.3723011087337094,
"grad_norm": 1.375,
"learning_rate": 0.0004990230251762418,
"loss": 5.6234,
"mean_token_accuracy": 0.15602398663759232,
"num_tokens": 8308515.0,
"step": 4785
},
{
"entropy": 5.635280561447144,
"epoch": 0.37269013810542695,
"grad_norm": 1.125,
"learning_rate": 0.0004990204434893713,
"loss": 5.5632,
"mean_token_accuracy": 0.16311416923999786,
"num_tokens": 8317260.0,
"step": 4790
},
{
"entropy": 5.789914894104004,
"epoch": 0.37307916747714454,
"grad_norm": 1.2265625,
"learning_rate": 0.0004990178584033474,
"loss": 5.7111,
"mean_token_accuracy": 0.14300042688846587,
"num_tokens": 8325887.0,
"step": 4795
},
{
"entropy": 5.82746148109436,
"epoch": 0.37346819684886207,
"grad_norm": 1.25,
"learning_rate": 0.0004990152699182089,
"loss": 5.7261,
"mean_token_accuracy": 0.13842646032571793,
"num_tokens": 8333974.0,
"step": 4800
},
{
"entropy": 5.683282423019409,
"epoch": 0.37385722622057965,
"grad_norm": 1.34375,
"learning_rate": 0.0004990126780339953,
"loss": 5.6969,
"mean_token_accuracy": 0.1456607311964035,
"num_tokens": 8343199.0,
"step": 4805
},
{
"entropy": 5.660903835296631,
"epoch": 0.37424625559229724,
"grad_norm": 1.21875,
"learning_rate": 0.0004990100827507459,
"loss": 5.4878,
"mean_token_accuracy": 0.15968211442232133,
"num_tokens": 8351202.0,
"step": 4810
},
{
"entropy": 5.794419288635254,
"epoch": 0.37463528496401477,
"grad_norm": 1.203125,
"learning_rate": 0.0004990074840684999,
"loss": 5.8265,
"mean_token_accuracy": 0.14525216594338416,
"num_tokens": 8360011.0,
"step": 4815
},
{
"entropy": 5.72889552116394,
"epoch": 0.37502431433573236,
"grad_norm": 1.3515625,
"learning_rate": 0.0004990048819872969,
"loss": 5.6967,
"mean_token_accuracy": 0.15121547728776932,
"num_tokens": 8367874.0,
"step": 4820
},
{
"entropy": 5.736692285537719,
"epoch": 0.3754133437074499,
"grad_norm": 1.328125,
"learning_rate": 0.0004990022765071765,
"loss": 5.6471,
"mean_token_accuracy": 0.15181145966053008,
"num_tokens": 8376708.0,
"step": 4825
},
{
"entropy": 5.776809597015381,
"epoch": 0.3758023730791675,
"grad_norm": 1.328125,
"learning_rate": 0.000498999667628178,
"loss": 5.6923,
"mean_token_accuracy": 0.15238475576043128,
"num_tokens": 8385467.0,
"step": 4830
},
{
"entropy": 5.693451023101806,
"epoch": 0.37619140245088506,
"grad_norm": 1.265625,
"learning_rate": 0.0004989970553503411,
"loss": 5.6671,
"mean_token_accuracy": 0.15169030725955962,
"num_tokens": 8393574.0,
"step": 4835
},
{
"entropy": 5.698660659790039,
"epoch": 0.3765804318226026,
"grad_norm": 1.2109375,
"learning_rate": 0.0004989944396737054,
"loss": 5.7057,
"mean_token_accuracy": 0.15399083346128464,
"num_tokens": 8402857.0,
"step": 4840
},
{
"entropy": 5.71474986076355,
"epoch": 0.3769694611943202,
"grad_norm": 1.28125,
"learning_rate": 0.0004989918205983106,
"loss": 5.6966,
"mean_token_accuracy": 0.16057430803775788,
"num_tokens": 8411728.0,
"step": 4845
},
{
"entropy": 5.776951026916504,
"epoch": 0.37735849056603776,
"grad_norm": 1.28125,
"learning_rate": 0.0004989891981241964,
"loss": 5.6816,
"mean_token_accuracy": 0.15508509129285813,
"num_tokens": 8420307.0,
"step": 4850
},
{
"entropy": 5.795963382720947,
"epoch": 0.3777475199377553,
"grad_norm": 1.2890625,
"learning_rate": 0.0004989865722514027,
"loss": 5.6859,
"mean_token_accuracy": 0.14701005667448044,
"num_tokens": 8429442.0,
"step": 4855
},
{
"entropy": 5.692721891403198,
"epoch": 0.3781365493094729,
"grad_norm": 1.203125,
"learning_rate": 0.0004989839429799692,
"loss": 5.5925,
"mean_token_accuracy": 0.15444754511117936,
"num_tokens": 8437945.0,
"step": 4860
},
{
"entropy": 5.69090576171875,
"epoch": 0.3785255786811904,
"grad_norm": 1.28125,
"learning_rate": 0.000498981310309936,
"loss": 5.5799,
"mean_token_accuracy": 0.1519486203789711,
"num_tokens": 8445743.0,
"step": 4865
},
{
"entropy": 5.673541116714477,
"epoch": 0.378914608052908,
"grad_norm": 1.28125,
"learning_rate": 0.0004989786742413428,
"loss": 5.6679,
"mean_token_accuracy": 0.14914161562919617,
"num_tokens": 8454091.0,
"step": 4870
},
{
"entropy": 5.828792095184326,
"epoch": 0.3793036374246256,
"grad_norm": 1.1953125,
"learning_rate": 0.0004989760347742298,
"loss": 5.6642,
"mean_token_accuracy": 0.148113676905632,
"num_tokens": 8461992.0,
"step": 4875
},
{
"entropy": 5.695930051803589,
"epoch": 0.3796926667963431,
"grad_norm": 1.2109375,
"learning_rate": 0.0004989733919086369,
"loss": 5.608,
"mean_token_accuracy": 0.1540228471159935,
"num_tokens": 8470730.0,
"step": 4880
},
{
"entropy": 5.694664525985718,
"epoch": 0.3800816961680607,
"grad_norm": 1.21875,
"learning_rate": 0.0004989707456446043,
"loss": 5.6546,
"mean_token_accuracy": 0.15131904631853105,
"num_tokens": 8479947.0,
"step": 4885
},
{
"entropy": 5.857856941223145,
"epoch": 0.38047072553977823,
"grad_norm": 1.2890625,
"learning_rate": 0.0004989680959821721,
"loss": 5.7356,
"mean_token_accuracy": 0.15022482126951217,
"num_tokens": 8488237.0,
"step": 4890
},
{
"entropy": 5.756908416748047,
"epoch": 0.3808597549114958,
"grad_norm": 1.203125,
"learning_rate": 0.0004989654429213805,
"loss": 5.5988,
"mean_token_accuracy": 0.15564934015274048,
"num_tokens": 8497146.0,
"step": 4895
},
{
"entropy": 5.723071098327637,
"epoch": 0.3812487842832134,
"grad_norm": 1.203125,
"learning_rate": 0.0004989627864622699,
"loss": 5.6833,
"mean_token_accuracy": 0.15712068229913712,
"num_tokens": 8505736.0,
"step": 4900
},
{
"entropy": 5.77219820022583,
"epoch": 0.38163781365493094,
"grad_norm": 1.234375,
"learning_rate": 0.0004989601266048804,
"loss": 5.6944,
"mean_token_accuracy": 0.15084800571203233,
"num_tokens": 8515368.0,
"step": 4905
},
{
"entropy": 5.703309774398804,
"epoch": 0.3820268430266485,
"grad_norm": 1.28125,
"learning_rate": 0.0004989574633492525,
"loss": 5.6671,
"mean_token_accuracy": 0.15129629224538804,
"num_tokens": 8524141.0,
"step": 4910
},
{
"entropy": 5.736749649047852,
"epoch": 0.38241587239836605,
"grad_norm": 1.21875,
"learning_rate": 0.0004989547966954266,
"loss": 5.5902,
"mean_token_accuracy": 0.15632274895906448,
"num_tokens": 8533116.0,
"step": 4915
},
{
"entropy": 5.701038837432861,
"epoch": 0.38280490177008364,
"grad_norm": 1.3046875,
"learning_rate": 0.000498952126643443,
"loss": 5.7171,
"mean_token_accuracy": 0.1558266080915928,
"num_tokens": 8541919.0,
"step": 4920
},
{
"entropy": 5.668884658813477,
"epoch": 0.3831939311418012,
"grad_norm": 1.234375,
"learning_rate": 0.0004989494531933424,
"loss": 5.6009,
"mean_token_accuracy": 0.15845879167318344,
"num_tokens": 8550372.0,
"step": 4925
},
{
"entropy": 5.705226564407349,
"epoch": 0.38358296051351876,
"grad_norm": 1.25,
"learning_rate": 0.0004989467763451652,
"loss": 5.6912,
"mean_token_accuracy": 0.15215400755405425,
"num_tokens": 8559246.0,
"step": 4930
},
{
"entropy": 5.691767168045044,
"epoch": 0.38397198988523634,
"grad_norm": 1.3046875,
"learning_rate": 0.0004989440960989523,
"loss": 5.4685,
"mean_token_accuracy": 0.16003949642181398,
"num_tokens": 8567005.0,
"step": 4935
},
{
"entropy": 5.635836219787597,
"epoch": 0.3843610192569539,
"grad_norm": 1.2109375,
"learning_rate": 0.000498941412454744,
"loss": 5.6521,
"mean_token_accuracy": 0.1453227147459984,
"num_tokens": 8575399.0,
"step": 4940
},
{
"entropy": 5.811139011383057,
"epoch": 0.38475004862867146,
"grad_norm": 1.265625,
"learning_rate": 0.0004989387254125813,
"loss": 5.7291,
"mean_token_accuracy": 0.14650810956954957,
"num_tokens": 8583994.0,
"step": 4945
},
{
"entropy": 5.760225105285644,
"epoch": 0.38513907800038905,
"grad_norm": 1.21875,
"learning_rate": 0.0004989360349725049,
"loss": 5.6496,
"mean_token_accuracy": 0.15658905506134033,
"num_tokens": 8593446.0,
"step": 4950
},
{
"entropy": 5.70222282409668,
"epoch": 0.3855281073721066,
"grad_norm": 1.28125,
"learning_rate": 0.0004989333411345555,
"loss": 5.6596,
"mean_token_accuracy": 0.15538214892148972,
"num_tokens": 8602821.0,
"step": 4955
},
{
"entropy": 5.706765222549438,
"epoch": 0.38591713674382416,
"grad_norm": 1.2890625,
"learning_rate": 0.0004989306438987742,
"loss": 5.7363,
"mean_token_accuracy": 0.15210612267255783,
"num_tokens": 8611831.0,
"step": 4960
},
{
"entropy": 5.772955226898193,
"epoch": 0.38630616611554175,
"grad_norm": 1.34375,
"learning_rate": 0.0004989279432652018,
"loss": 5.6422,
"mean_token_accuracy": 0.1529601663351059,
"num_tokens": 8620986.0,
"step": 4965
},
{
"entropy": 5.727062749862671,
"epoch": 0.3866951954872593,
"grad_norm": 1.28125,
"learning_rate": 0.0004989252392338791,
"loss": 5.5982,
"mean_token_accuracy": 0.15300055742263793,
"num_tokens": 8629822.0,
"step": 4970
},
{
"entropy": 5.673508405685425,
"epoch": 0.38708422485897687,
"grad_norm": 1.2890625,
"learning_rate": 0.0004989225318048475,
"loss": 5.6862,
"mean_token_accuracy": 0.1478674292564392,
"num_tokens": 8637679.0,
"step": 4975
},
{
"entropy": 5.793036079406738,
"epoch": 0.3874732542306944,
"grad_norm": 1.609375,
"learning_rate": 0.0004989198209781478,
"loss": 5.721,
"mean_token_accuracy": 0.14822040498256683,
"num_tokens": 8646705.0,
"step": 4980
},
{
"entropy": 5.799608898162842,
"epoch": 0.387862283602412,
"grad_norm": 1.234375,
"learning_rate": 0.0004989171067538211,
"loss": 5.6281,
"mean_token_accuracy": 0.15128099098801612,
"num_tokens": 8655599.0,
"step": 4985
},
{
"entropy": 5.635348701477051,
"epoch": 0.38825131297412957,
"grad_norm": 1.1796875,
"learning_rate": 0.000498914389131909,
"loss": 5.6514,
"mean_token_accuracy": 0.15804774016141893,
"num_tokens": 8665025.0,
"step": 4990
},
{
"entropy": 5.719934034347534,
"epoch": 0.3886403423458471,
"grad_norm": 1.4296875,
"learning_rate": 0.0004989116681124523,
"loss": 5.5337,
"mean_token_accuracy": 0.15749720931053163,
"num_tokens": 8672994.0,
"step": 4995
},
{
"entropy": 5.669281339645385,
"epoch": 0.3890293717175647,
"grad_norm": 1.2578125,
"learning_rate": 0.0004989089436954924,
"loss": 5.6883,
"mean_token_accuracy": 0.15234919488430024,
"num_tokens": 8682457.0,
"step": 5000
},
{
"entropy": 5.705449628829956,
"epoch": 0.3894184010892822,
"grad_norm": 1.28125,
"learning_rate": 0.0004989062158810706,
"loss": 5.662,
"mean_token_accuracy": 0.15314388871192933,
"num_tokens": 8690888.0,
"step": 5005
},
{
"entropy": 5.873299598693848,
"epoch": 0.3898074304609998,
"grad_norm": 1.3125,
"learning_rate": 0.0004989034846692284,
"loss": 5.7437,
"mean_token_accuracy": 0.15064525082707406,
"num_tokens": 8699632.0,
"step": 5010
},
{
"entropy": 5.6710282325744625,
"epoch": 0.3901964598327174,
"grad_norm": 1.3515625,
"learning_rate": 0.0004989007500600073,
"loss": 5.5879,
"mean_token_accuracy": 0.1571871817111969,
"num_tokens": 8707955.0,
"step": 5015
},
{
"entropy": 5.7133081436157225,
"epoch": 0.3905854892044349,
"grad_norm": 1.3984375,
"learning_rate": 0.0004988980120534486,
"loss": 5.737,
"mean_token_accuracy": 0.15538298040628434,
"num_tokens": 8716103.0,
"step": 5020
},
{
"entropy": 5.751881408691406,
"epoch": 0.3909745185761525,
"grad_norm": 1.3046875,
"learning_rate": 0.000498895270649594,
"loss": 5.5882,
"mean_token_accuracy": 0.16275588870048524,
"num_tokens": 8724499.0,
"step": 5025
},
{
"entropy": 5.674628114700317,
"epoch": 0.39136354794787004,
"grad_norm": 1.296875,
"learning_rate": 0.000498892525848485,
"loss": 5.6015,
"mean_token_accuracy": 0.15894443094730376,
"num_tokens": 8732829.0,
"step": 5030
},
{
"entropy": 5.611494064331055,
"epoch": 0.3917525773195876,
"grad_norm": 1.2734375,
"learning_rate": 0.0004988897776501633,
"loss": 5.465,
"mean_token_accuracy": 0.15986961871385574,
"num_tokens": 8741617.0,
"step": 5035
},
{
"entropy": 5.659196472167968,
"epoch": 0.3921416066913052,
"grad_norm": 1.1875,
"learning_rate": 0.0004988870260546705,
"loss": 5.467,
"mean_token_accuracy": 0.16652076989412307,
"num_tokens": 8750146.0,
"step": 5040
},
{
"entropy": 5.657342576980591,
"epoch": 0.39253063606302274,
"grad_norm": 1.2578125,
"learning_rate": 0.0004988842710620486,
"loss": 5.6448,
"mean_token_accuracy": 0.15352269858121873,
"num_tokens": 8758744.0,
"step": 5045
},
{
"entropy": 5.6410760402679445,
"epoch": 0.39291966543474033,
"grad_norm": 1.2265625,
"learning_rate": 0.0004988815126723391,
"loss": 5.6622,
"mean_token_accuracy": 0.15248211920261384,
"num_tokens": 8768219.0,
"step": 5050
},
{
"entropy": 5.785637426376343,
"epoch": 0.3933086948064579,
"grad_norm": 1.2421875,
"learning_rate": 0.0004988787508855841,
"loss": 5.6576,
"mean_token_accuracy": 0.1519287884235382,
"num_tokens": 8777047.0,
"step": 5055
},
{
"entropy": 5.642466020584107,
"epoch": 0.39369772417817545,
"grad_norm": 1.2734375,
"learning_rate": 0.0004988759857018253,
"loss": 5.5433,
"mean_token_accuracy": 0.15469427406787872,
"num_tokens": 8785918.0,
"step": 5060
},
{
"entropy": 5.631013011932373,
"epoch": 0.39408675354989303,
"grad_norm": 1.3046875,
"learning_rate": 0.0004988732171211048,
"loss": 5.6233,
"mean_token_accuracy": 0.14678166359663009,
"num_tokens": 8794253.0,
"step": 5065
},
{
"entropy": 5.76476469039917,
"epoch": 0.39447578292161056,
"grad_norm": 1.2734375,
"learning_rate": 0.0004988704451434644,
"loss": 5.6719,
"mean_token_accuracy": 0.16016739159822463,
"num_tokens": 8802352.0,
"step": 5070
},
{
"entropy": 5.78267993927002,
"epoch": 0.39486481229332815,
"grad_norm": 1.359375,
"learning_rate": 0.0004988676697689465,
"loss": 5.6242,
"mean_token_accuracy": 0.15009063333272935,
"num_tokens": 8810956.0,
"step": 5075
},
{
"entropy": 5.654299974441528,
"epoch": 0.39525384166504574,
"grad_norm": 1.2265625,
"learning_rate": 0.000498864890997593,
"loss": 5.5622,
"mean_token_accuracy": 0.1525967687368393,
"num_tokens": 8819062.0,
"step": 5080
},
{
"entropy": 5.7077337265014645,
"epoch": 0.39564287103676327,
"grad_norm": 1.3671875,
"learning_rate": 0.0004988621088294461,
"loss": 5.6572,
"mean_token_accuracy": 0.1502836376428604,
"num_tokens": 8827470.0,
"step": 5085
},
{
"entropy": 5.721915864944458,
"epoch": 0.39603190040848085,
"grad_norm": 1.3671875,
"learning_rate": 0.000498859323264548,
"loss": 5.6916,
"mean_token_accuracy": 0.15031601786613463,
"num_tokens": 8837003.0,
"step": 5090
},
{
"entropy": 5.713486862182617,
"epoch": 0.3964209297801984,
"grad_norm": 1.3203125,
"learning_rate": 0.000498856534302941,
"loss": 5.697,
"mean_token_accuracy": 0.14879803881049156,
"num_tokens": 8846605.0,
"step": 5095
},
{
"entropy": 5.79709038734436,
"epoch": 0.39680995915191597,
"grad_norm": 1.3046875,
"learning_rate": 0.0004988537419446673,
"loss": 5.7335,
"mean_token_accuracy": 0.14815959110856056,
"num_tokens": 8855232.0,
"step": 5100
},
{
"entropy": 5.755407381057739,
"epoch": 0.39719898852363356,
"grad_norm": 1.2734375,
"learning_rate": 0.0004988509461897694,
"loss": 5.6102,
"mean_token_accuracy": 0.16033030301332474,
"num_tokens": 8864176.0,
"step": 5105
},
{
"entropy": 5.733327960968017,
"epoch": 0.3975880178953511,
"grad_norm": 1.21875,
"learning_rate": 0.0004988481470382897,
"loss": 5.6897,
"mean_token_accuracy": 0.156675386428833,
"num_tokens": 8872671.0,
"step": 5110
},
{
"entropy": 5.610649251937867,
"epoch": 0.3979770472670687,
"grad_norm": 1.3203125,
"learning_rate": 0.0004988453444902708,
"loss": 5.6034,
"mean_token_accuracy": 0.1590520143508911,
"num_tokens": 8881316.0,
"step": 5115
},
{
"entropy": 5.760883426666259,
"epoch": 0.3983660766387862,
"grad_norm": 1.375,
"learning_rate": 0.0004988425385457549,
"loss": 5.719,
"mean_token_accuracy": 0.15436383709311485,
"num_tokens": 8890114.0,
"step": 5120
},
{
"entropy": 5.7600020408630375,
"epoch": 0.3987551060105038,
"grad_norm": 1.6171875,
"learning_rate": 0.0004988397292047848,
"loss": 5.6282,
"mean_token_accuracy": 0.15803250372409822,
"num_tokens": 8899598.0,
"step": 5125
},
{
"entropy": 5.609093189239502,
"epoch": 0.3991441353822214,
"grad_norm": 1.359375,
"learning_rate": 0.0004988369164674032,
"loss": 5.5068,
"mean_token_accuracy": 0.15966605246067048,
"num_tokens": 8908244.0,
"step": 5130
},
{
"entropy": 5.688562393188477,
"epoch": 0.3995331647539389,
"grad_norm": 1.3125,
"learning_rate": 0.0004988341003336526,
"loss": 5.5515,
"mean_token_accuracy": 0.1559025228023529,
"num_tokens": 8916280.0,
"step": 5135
},
{
"entropy": 5.735493183135986,
"epoch": 0.3999221941256565,
"grad_norm": 1.34375,
"learning_rate": 0.0004988312808035757,
"loss": 5.5901,
"mean_token_accuracy": 0.15177929475903512,
"num_tokens": 8923947.0,
"step": 5140
},
{
"entropy": 5.653040266036987,
"epoch": 0.400311223497374,
"grad_norm": 1.4296875,
"learning_rate": 0.0004988284578772155,
"loss": 5.6619,
"mean_token_accuracy": 0.14781947880983354,
"num_tokens": 8933374.0,
"step": 5145
},
{
"entropy": 5.672594928741455,
"epoch": 0.4007002528690916,
"grad_norm": 1.328125,
"learning_rate": 0.0004988256315546146,
"loss": 5.5607,
"mean_token_accuracy": 0.15781172960996628,
"num_tokens": 8941757.0,
"step": 5150
},
{
"entropy": 5.663328552246094,
"epoch": 0.4010892822408092,
"grad_norm": 1.265625,
"learning_rate": 0.0004988228018358161,
"loss": 5.4872,
"mean_token_accuracy": 0.15782520771026612,
"num_tokens": 8950128.0,
"step": 5155
},
{
"entropy": 5.73198447227478,
"epoch": 0.40147831161252673,
"grad_norm": 1.3828125,
"learning_rate": 0.0004988199687208628,
"loss": 5.6454,
"mean_token_accuracy": 0.14845688492059708,
"num_tokens": 8958939.0,
"step": 5160
},
{
"entropy": 5.774505949020385,
"epoch": 0.4018673409842443,
"grad_norm": 1.171875,
"learning_rate": 0.0004988171322097977,
"loss": 5.604,
"mean_token_accuracy": 0.15679994523525237,
"num_tokens": 8967519.0,
"step": 5165
},
{
"entropy": 5.595438528060913,
"epoch": 0.4022563703559619,
"grad_norm": 1.3125,
"learning_rate": 0.0004988142923026638,
"loss": 5.4646,
"mean_token_accuracy": 0.1620745465159416,
"num_tokens": 8976215.0,
"step": 5170
},
{
"entropy": 5.698001861572266,
"epoch": 0.40264539972767943,
"grad_norm": 1.359375,
"learning_rate": 0.0004988114489995044,
"loss": 5.7052,
"mean_token_accuracy": 0.14649326354265213,
"num_tokens": 8984060.0,
"step": 5175
},
{
"entropy": 5.712637948989868,
"epoch": 0.403034429099397,
"grad_norm": 1.28125,
"learning_rate": 0.0004988086023003624,
"loss": 5.6301,
"mean_token_accuracy": 0.15002061128616334,
"num_tokens": 8992301.0,
"step": 5180
},
{
"entropy": 5.742618560791016,
"epoch": 0.40342345847111455,
"grad_norm": 1.453125,
"learning_rate": 0.0004988057522052811,
"loss": 5.6659,
"mean_token_accuracy": 0.15351343005895615,
"num_tokens": 9000517.0,
"step": 5185
},
{
"entropy": 5.68753137588501,
"epoch": 0.40381248784283214,
"grad_norm": 1.4296875,
"learning_rate": 0.0004988028987143037,
"loss": 5.6106,
"mean_token_accuracy": 0.15389739722013474,
"num_tokens": 9008826.0,
"step": 5190
},
{
"entropy": 5.725468111038208,
"epoch": 0.4042015172145497,
"grad_norm": 1.2734375,
"learning_rate": 0.0004988000418274736,
"loss": 5.6164,
"mean_token_accuracy": 0.14825106114149095,
"num_tokens": 9017952.0,
"step": 5195
},
{
"entropy": 5.672607660293579,
"epoch": 0.40459054658626725,
"grad_norm": 1.234375,
"learning_rate": 0.0004987971815448341,
"loss": 5.6115,
"mean_token_accuracy": 0.15911526530981063,
"num_tokens": 9026688.0,
"step": 5200
},
{
"entropy": 5.678609228134155,
"epoch": 0.40497957595798484,
"grad_norm": 1.4296875,
"learning_rate": 0.0004987943178664285,
"loss": 5.5947,
"mean_token_accuracy": 0.1578623317182064,
"num_tokens": 9035761.0,
"step": 5205
},
{
"entropy": 5.737362957000732,
"epoch": 0.40536860532970237,
"grad_norm": 1.2890625,
"learning_rate": 0.0004987914507923004,
"loss": 5.7061,
"mean_token_accuracy": 0.1476090058684349,
"num_tokens": 9044510.0,
"step": 5210
},
{
"entropy": 5.624804878234864,
"epoch": 0.40575763470141996,
"grad_norm": 1.328125,
"learning_rate": 0.0004987885803224931,
"loss": 5.5507,
"mean_token_accuracy": 0.1544179081916809,
"num_tokens": 9053031.0,
"step": 5215
},
{
"entropy": 5.6066936492919925,
"epoch": 0.40614666407313754,
"grad_norm": 1.421875,
"learning_rate": 0.0004987857064570505,
"loss": 5.5451,
"mean_token_accuracy": 0.16021619886159896,
"num_tokens": 9061580.0,
"step": 5220
},
{
"entropy": 5.718929815292358,
"epoch": 0.4065356934448551,
"grad_norm": 1.2734375,
"learning_rate": 0.0004987828291960158,
"loss": 5.6634,
"mean_token_accuracy": 0.1563795655965805,
"num_tokens": 9069741.0,
"step": 5225
},
{
"entropy": 5.811468648910522,
"epoch": 0.40692472281657266,
"grad_norm": 1.25,
"learning_rate": 0.000498779948539433,
"loss": 5.6813,
"mean_token_accuracy": 0.143922408670187,
"num_tokens": 9079214.0,
"step": 5230
},
{
"entropy": 5.773811149597168,
"epoch": 0.4073137521882902,
"grad_norm": 1.1875,
"learning_rate": 0.0004987770644873455,
"loss": 5.6583,
"mean_token_accuracy": 0.15696950107812882,
"num_tokens": 9088201.0,
"step": 5235
},
{
"entropy": 5.716661548614502,
"epoch": 0.4077027815600078,
"grad_norm": 1.3359375,
"learning_rate": 0.0004987741770397974,
"loss": 5.6668,
"mean_token_accuracy": 0.14920853078365326,
"num_tokens": 9096829.0,
"step": 5240
},
{
"entropy": 5.703339910507202,
"epoch": 0.40809181093172536,
"grad_norm": 1.34375,
"learning_rate": 0.0004987712861968323,
"loss": 5.6269,
"mean_token_accuracy": 0.15427666306495666,
"num_tokens": 9105515.0,
"step": 5245
},
{
"entropy": 5.777084398269653,
"epoch": 0.4084808403034429,
"grad_norm": 1.4921875,
"learning_rate": 0.000498768391958494,
"loss": 5.6073,
"mean_token_accuracy": 0.15668865889310837,
"num_tokens": 9113614.0,
"step": 5250
},
{
"entropy": 5.712219905853272,
"epoch": 0.4088698696751605,
"grad_norm": 1.2734375,
"learning_rate": 0.0004987654943248266,
"loss": 5.697,
"mean_token_accuracy": 0.14341746270656586,
"num_tokens": 9122337.0,
"step": 5255
},
{
"entropy": 5.630028343200683,
"epoch": 0.40925889904687807,
"grad_norm": 1.421875,
"learning_rate": 0.0004987625932958739,
"loss": 5.5779,
"mean_token_accuracy": 0.15674022287130357,
"num_tokens": 9130417.0,
"step": 5260
},
{
"entropy": 5.7025354385375975,
"epoch": 0.4096479284185956,
"grad_norm": 1.2265625,
"learning_rate": 0.0004987596888716801,
"loss": 5.6483,
"mean_token_accuracy": 0.14859899282455444,
"num_tokens": 9139317.0,
"step": 5265
},
{
"entropy": 5.726134777069092,
"epoch": 0.4100369577903132,
"grad_norm": 1.3125,
"learning_rate": 0.0004987567810522892,
"loss": 5.626,
"mean_token_accuracy": 0.15348759293556213,
"num_tokens": 9147808.0,
"step": 5270
},
{
"entropy": 5.818262100219727,
"epoch": 0.4104259871620307,
"grad_norm": 1.1953125,
"learning_rate": 0.0004987538698377451,
"loss": 5.7377,
"mean_token_accuracy": 0.14077338203787804,
"num_tokens": 9157117.0,
"step": 5275
},
{
"entropy": 5.736328840255737,
"epoch": 0.4108150165337483,
"grad_norm": 1.25,
"learning_rate": 0.0004987509552280924,
"loss": 5.573,
"mean_token_accuracy": 0.15244077891111374,
"num_tokens": 9165297.0,
"step": 5280
},
{
"entropy": 5.708140563964844,
"epoch": 0.4112040459054659,
"grad_norm": 1.2578125,
"learning_rate": 0.000498748037223375,
"loss": 5.7137,
"mean_token_accuracy": 0.151723912358284,
"num_tokens": 9172945.0,
"step": 5285
},
{
"entropy": 5.745821762084961,
"epoch": 0.4115930752771834,
"grad_norm": 1.3359375,
"learning_rate": 0.0004987451158236372,
"loss": 5.6428,
"mean_token_accuracy": 0.15701709389686586,
"num_tokens": 9181445.0,
"step": 5290
},
{
"entropy": 5.605129194259644,
"epoch": 0.411982104648901,
"grad_norm": 1.34375,
"learning_rate": 0.0004987421910289234,
"loss": 5.4712,
"mean_token_accuracy": 0.1588175266981125,
"num_tokens": 9189535.0,
"step": 5295
},
{
"entropy": 5.649341869354248,
"epoch": 0.41237113402061853,
"grad_norm": 1.3203125,
"learning_rate": 0.0004987392628392781,
"loss": 5.6881,
"mean_token_accuracy": 0.15033083707094191,
"num_tokens": 9198328.0,
"step": 5300
},
{
"entropy": 5.740281772613526,
"epoch": 0.4127601633923361,
"grad_norm": 1.3359375,
"learning_rate": 0.0004987363312547456,
"loss": 5.6035,
"mean_token_accuracy": 0.1550357922911644,
"num_tokens": 9206840.0,
"step": 5305
},
{
"entropy": 5.686477708816528,
"epoch": 0.4131491927640537,
"grad_norm": 1.375,
"learning_rate": 0.0004987333962753703,
"loss": 5.5073,
"mean_token_accuracy": 0.1557457096874714,
"num_tokens": 9215155.0,
"step": 5310
},
{
"entropy": 5.689324522018433,
"epoch": 0.41353822213577124,
"grad_norm": 1.3125,
"learning_rate": 0.0004987304579011967,
"loss": 5.6741,
"mean_token_accuracy": 0.1526474580168724,
"num_tokens": 9224426.0,
"step": 5315
},
{
"entropy": 5.768082618713379,
"epoch": 0.4139272515074888,
"grad_norm": 1.421875,
"learning_rate": 0.0004987275161322697,
"loss": 5.6794,
"mean_token_accuracy": 0.1533222734928131,
"num_tokens": 9233785.0,
"step": 5320
},
{
"entropy": 5.695648050308227,
"epoch": 0.41431628087920636,
"grad_norm": 1.28125,
"learning_rate": 0.0004987245709686337,
"loss": 5.6307,
"mean_token_accuracy": 0.15449224412441254,
"num_tokens": 9242168.0,
"step": 5325
},
{
"entropy": 5.718257284164428,
"epoch": 0.41470531025092394,
"grad_norm": 1.25,
"learning_rate": 0.0004987216224103334,
"loss": 5.6645,
"mean_token_accuracy": 0.15255628675222396,
"num_tokens": 9250862.0,
"step": 5330
},
{
"entropy": 5.7047186374664305,
"epoch": 0.41509433962264153,
"grad_norm": 1.265625,
"learning_rate": 0.0004987186704574136,
"loss": 5.6741,
"mean_token_accuracy": 0.14798310920596122,
"num_tokens": 9259957.0,
"step": 5335
},
{
"entropy": 5.61541223526001,
"epoch": 0.41548336899435906,
"grad_norm": 1.3203125,
"learning_rate": 0.0004987157151099189,
"loss": 5.5546,
"mean_token_accuracy": 0.15545115619897842,
"num_tokens": 9268590.0,
"step": 5340
},
{
"entropy": 5.702762413024902,
"epoch": 0.41587239836607665,
"grad_norm": 1.3125,
"learning_rate": 0.0004987127563678945,
"loss": 5.6005,
"mean_token_accuracy": 0.15641995817422866,
"num_tokens": 9276872.0,
"step": 5345
},
{
"entropy": 5.679139614105225,
"epoch": 0.4162614277377942,
"grad_norm": 1.296875,
"learning_rate": 0.0004987097942313852,
"loss": 5.5116,
"mean_token_accuracy": 0.15718710869550706,
"num_tokens": 9284556.0,
"step": 5350
},
{
"entropy": 5.651728963851928,
"epoch": 0.41665045710951176,
"grad_norm": 1.28125,
"learning_rate": 0.0004987068287004355,
"loss": 5.6732,
"mean_token_accuracy": 0.15081616044044494,
"num_tokens": 9293315.0,
"step": 5355
},
{
"entropy": 5.773790979385376,
"epoch": 0.41703948648122935,
"grad_norm": 1.3046875,
"learning_rate": 0.0004987038597750909,
"loss": 5.6363,
"mean_token_accuracy": 0.15579553544521332,
"num_tokens": 9302923.0,
"step": 5360
},
{
"entropy": 5.722340250015259,
"epoch": 0.4174285158529469,
"grad_norm": 1.4140625,
"learning_rate": 0.0004987008874553964,
"loss": 5.5691,
"mean_token_accuracy": 0.1573861852288246,
"num_tokens": 9311388.0,
"step": 5365
},
{
"entropy": 5.6082155227661135,
"epoch": 0.41781754522466447,
"grad_norm": 1.3671875,
"learning_rate": 0.0004986979117413969,
"loss": 5.5166,
"mean_token_accuracy": 0.15783468931913375,
"num_tokens": 9319880.0,
"step": 5370
},
{
"entropy": 5.66514778137207,
"epoch": 0.41820657459638205,
"grad_norm": 2.390625,
"learning_rate": 0.0004986949326331376,
"loss": 5.6584,
"mean_token_accuracy": 0.15248090997338296,
"num_tokens": 9328374.0,
"step": 5375
},
{
"entropy": 5.7879310131073,
"epoch": 0.4185956039680996,
"grad_norm": 1.3984375,
"learning_rate": 0.0004986919501306638,
"loss": 5.6027,
"mean_token_accuracy": 0.16203720420598983,
"num_tokens": 9336428.0,
"step": 5380
},
{
"entropy": 5.629948520660401,
"epoch": 0.41898463333981717,
"grad_norm": 1.1640625,
"learning_rate": 0.0004986889642340207,
"loss": 5.7067,
"mean_token_accuracy": 0.15284340232610702,
"num_tokens": 9345794.0,
"step": 5385
},
{
"entropy": 5.776071500778198,
"epoch": 0.4193736627115347,
"grad_norm": 1.28125,
"learning_rate": 0.0004986859749432536,
"loss": 5.6636,
"mean_token_accuracy": 0.1524375021457672,
"num_tokens": 9354724.0,
"step": 5390
},
{
"entropy": 5.768964719772339,
"epoch": 0.4197626920832523,
"grad_norm": 1.328125,
"learning_rate": 0.0004986829822584078,
"loss": 5.6802,
"mean_token_accuracy": 0.1479766398668289,
"num_tokens": 9364008.0,
"step": 5395
},
{
"entropy": 5.72349443435669,
"epoch": 0.4201517214549699,
"grad_norm": 1.3125,
"learning_rate": 0.0004986799861795289,
"loss": 5.6363,
"mean_token_accuracy": 0.16576656699180603,
"num_tokens": 9372409.0,
"step": 5400
},
{
"entropy": 5.697003746032715,
"epoch": 0.4205407508266874,
"grad_norm": 1.3984375,
"learning_rate": 0.0004986769867066622,
"loss": 5.579,
"mean_token_accuracy": 0.15202543288469314,
"num_tokens": 9381172.0,
"step": 5405
},
{
"entropy": 5.68998646736145,
"epoch": 0.420929780198405,
"grad_norm": 1.484375,
"learning_rate": 0.0004986739838398532,
"loss": 5.6024,
"mean_token_accuracy": 0.16151912212371827,
"num_tokens": 9389245.0,
"step": 5410
},
{
"entropy": 5.710152196884155,
"epoch": 0.4213188095701225,
"grad_norm": 1.390625,
"learning_rate": 0.0004986709775791475,
"loss": 5.6409,
"mean_token_accuracy": 0.15243250578641893,
"num_tokens": 9397954.0,
"step": 5415
},
{
"entropy": 5.68855242729187,
"epoch": 0.4217078389418401,
"grad_norm": 1.265625,
"learning_rate": 0.0004986679679245907,
"loss": 5.7307,
"mean_token_accuracy": 0.14909764230251313,
"num_tokens": 9406986.0,
"step": 5420
},
{
"entropy": 5.755280351638794,
"epoch": 0.4220968683135577,
"grad_norm": 1.3125,
"learning_rate": 0.0004986649548762286,
"loss": 5.6857,
"mean_token_accuracy": 0.15256578773260115,
"num_tokens": 9415163.0,
"step": 5425
},
{
"entropy": 5.672216081619263,
"epoch": 0.4224858976852752,
"grad_norm": 1.234375,
"learning_rate": 0.0004986619384341066,
"loss": 5.5171,
"mean_token_accuracy": 0.1674740955233574,
"num_tokens": 9423422.0,
"step": 5430
},
{
"entropy": 5.641525936126709,
"epoch": 0.4228749270569928,
"grad_norm": 1.2578125,
"learning_rate": 0.0004986589185982708,
"loss": 5.554,
"mean_token_accuracy": 0.1703364595770836,
"num_tokens": 9431349.0,
"step": 5435
},
{
"entropy": 5.704486656188965,
"epoch": 0.42326395642871034,
"grad_norm": 1.5,
"learning_rate": 0.0004986558953687671,
"loss": 5.6504,
"mean_token_accuracy": 0.1514353260397911,
"num_tokens": 9439715.0,
"step": 5440
},
{
"entropy": 5.736121320724488,
"epoch": 0.42365298580042793,
"grad_norm": 1.2421875,
"learning_rate": 0.0004986528687456409,
"loss": 5.6528,
"mean_token_accuracy": 0.15465957820415496,
"num_tokens": 9448216.0,
"step": 5445
},
{
"entropy": 5.739035606384277,
"epoch": 0.4240420151721455,
"grad_norm": 1.3203125,
"learning_rate": 0.0004986498387289384,
"loss": 5.5811,
"mean_token_accuracy": 0.1586304396390915,
"num_tokens": 9456346.0,
"step": 5450
},
{
"entropy": 5.70108847618103,
"epoch": 0.42443104454386305,
"grad_norm": 1.2890625,
"learning_rate": 0.0004986468053187058,
"loss": 5.6537,
"mean_token_accuracy": 0.1538468062877655,
"num_tokens": 9465476.0,
"step": 5455
},
{
"entropy": 5.770843505859375,
"epoch": 0.42482007391558063,
"grad_norm": 1.21875,
"learning_rate": 0.0004986437685149887,
"loss": 5.7147,
"mean_token_accuracy": 0.1504819832742214,
"num_tokens": 9474582.0,
"step": 5460
},
{
"entropy": 5.699989891052246,
"epoch": 0.4252091032872982,
"grad_norm": 1.265625,
"learning_rate": 0.0004986407283178334,
"loss": 5.5785,
"mean_token_accuracy": 0.14940586686134338,
"num_tokens": 9482801.0,
"step": 5465
},
{
"entropy": 5.649980115890503,
"epoch": 0.42559813265901575,
"grad_norm": 1.3828125,
"learning_rate": 0.0004986376847272861,
"loss": 5.6356,
"mean_token_accuracy": 0.15340728610754012,
"num_tokens": 9490894.0,
"step": 5470
},
{
"entropy": 5.674148035049439,
"epoch": 0.42598716203073334,
"grad_norm": 1.28125,
"learning_rate": 0.0004986346377433929,
"loss": 5.5886,
"mean_token_accuracy": 0.155898617208004,
"num_tokens": 9499414.0,
"step": 5475
},
{
"entropy": 5.6370621681213375,
"epoch": 0.42637619140245087,
"grad_norm": 1.2734375,
"learning_rate": 0.0004986315873661999,
"loss": 5.5502,
"mean_token_accuracy": 0.15451737493276596,
"num_tokens": 9508945.0,
"step": 5480
},
{
"entropy": 5.727960109710693,
"epoch": 0.42676522077416845,
"grad_norm": 1.2890625,
"learning_rate": 0.0004986285335957536,
"loss": 5.6277,
"mean_token_accuracy": 0.1591019555926323,
"num_tokens": 9517833.0,
"step": 5485
},
{
"entropy": 5.782824993133545,
"epoch": 0.42715425014588604,
"grad_norm": 1.328125,
"learning_rate": 0.0004986254764321002,
"loss": 5.768,
"mean_token_accuracy": 0.14558715224266053,
"num_tokens": 9526551.0,
"step": 5490
},
{
"entropy": 5.6884478569030765,
"epoch": 0.42754327951760357,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986224158752861,
"loss": 5.5169,
"mean_token_accuracy": 0.15997313857078552,
"num_tokens": 9535224.0,
"step": 5495
},
{
"entropy": 5.723999786376953,
"epoch": 0.42793230888932116,
"grad_norm": 1.2734375,
"learning_rate": 0.0004986193519253578,
"loss": 5.647,
"mean_token_accuracy": 0.14827475771307946,
"num_tokens": 9544466.0,
"step": 5500
},
{
"entropy": 5.652610635757446,
"epoch": 0.4283213382610387,
"grad_norm": 1.234375,
"learning_rate": 0.0004986162845823618,
"loss": 5.6416,
"mean_token_accuracy": 0.1531407952308655,
"num_tokens": 9553520.0,
"step": 5505
},
{
"entropy": 5.6986065864562985,
"epoch": 0.4287103676327563,
"grad_norm": 1.3203125,
"learning_rate": 0.0004986132138463446,
"loss": 5.5536,
"mean_token_accuracy": 0.1587995931506157,
"num_tokens": 9562053.0,
"step": 5510
},
{
"entropy": 5.650413608551025,
"epoch": 0.42909939700447386,
"grad_norm": 1.3125,
"learning_rate": 0.0004986101397173528,
"loss": 5.5361,
"mean_token_accuracy": 0.15939041525125502,
"num_tokens": 9570632.0,
"step": 5515
},
{
"entropy": 5.637784433364868,
"epoch": 0.4294884263761914,
"grad_norm": 1.28125,
"learning_rate": 0.000498607062195433,
"loss": 5.5628,
"mean_token_accuracy": 0.15918782651424407,
"num_tokens": 9578739.0,
"step": 5520
},
{
"entropy": 5.681663131713867,
"epoch": 0.429877455747909,
"grad_norm": 1.2109375,
"learning_rate": 0.000498603981280632,
"loss": 5.5821,
"mean_token_accuracy": 0.15625160336494445,
"num_tokens": 9587314.0,
"step": 5525
},
{
"entropy": 5.723372554779052,
"epoch": 0.4302664851196265,
"grad_norm": 1.40625,
"learning_rate": 0.0004986008969729964,
"loss": 5.6419,
"mean_token_accuracy": 0.15364105701446534,
"num_tokens": 9595954.0,
"step": 5530
},
{
"entropy": 5.693819379806518,
"epoch": 0.4306555144913441,
"grad_norm": 1.359375,
"learning_rate": 0.0004985978092725731,
"loss": 5.6581,
"mean_token_accuracy": 0.1521173059940338,
"num_tokens": 9604767.0,
"step": 5535
},
{
"entropy": 5.6503571510314945,
"epoch": 0.4310445438630617,
"grad_norm": 1.34375,
"learning_rate": 0.000498594718179409,
"loss": 5.5238,
"mean_token_accuracy": 0.15622634440660477,
"num_tokens": 9613173.0,
"step": 5540
},
{
"entropy": 5.711286640167236,
"epoch": 0.4314335732347792,
"grad_norm": 1.296875,
"learning_rate": 0.0004985916236935508,
"loss": 5.6142,
"mean_token_accuracy": 0.15553787499666213,
"num_tokens": 9622777.0,
"step": 5545
},
{
"entropy": 5.608004999160767,
"epoch": 0.4318226026064968,
"grad_norm": 1.359375,
"learning_rate": 0.0004985885258150458,
"loss": 5.548,
"mean_token_accuracy": 0.15261236429214478,
"num_tokens": 9631019.0,
"step": 5550
},
{
"entropy": 5.66903657913208,
"epoch": 0.43221163197821433,
"grad_norm": 1.1953125,
"learning_rate": 0.0004985854245439408,
"loss": 5.6201,
"mean_token_accuracy": 0.1510221004486084,
"num_tokens": 9639797.0,
"step": 5555
},
{
"entropy": 5.675737190246582,
"epoch": 0.4326006613499319,
"grad_norm": 1.3515625,
"learning_rate": 0.0004985823198802827,
"loss": 5.5886,
"mean_token_accuracy": 0.16301481425762177,
"num_tokens": 9648393.0,
"step": 5560
},
{
"entropy": 5.710528373718262,
"epoch": 0.4329896907216495,
"grad_norm": 1.4375,
"learning_rate": 0.0004985792118241188,
"loss": 5.5958,
"mean_token_accuracy": 0.15752308517694474,
"num_tokens": 9657260.0,
"step": 5565
},
{
"entropy": 5.668179178237915,
"epoch": 0.43337872009336703,
"grad_norm": 1.3359375,
"learning_rate": 0.0004985761003754961,
"loss": 5.6327,
"mean_token_accuracy": 0.16131462305784225,
"num_tokens": 9665883.0,
"step": 5570
},
{
"entropy": 5.726443815231323,
"epoch": 0.4337677494650846,
"grad_norm": 1.4140625,
"learning_rate": 0.0004985729855344622,
"loss": 5.6628,
"mean_token_accuracy": 0.15568624585866928,
"num_tokens": 9674100.0,
"step": 5575
},
{
"entropy": 5.669579696655274,
"epoch": 0.4341567788368022,
"grad_norm": 1.4296875,
"learning_rate": 0.000498569867301064,
"loss": 5.6411,
"mean_token_accuracy": 0.14948973059654236,
"num_tokens": 9683236.0,
"step": 5580
},
{
"entropy": 5.671548175811767,
"epoch": 0.43454580820851973,
"grad_norm": 1.3671875,
"learning_rate": 0.0004985667456753489,
"loss": 5.6118,
"mean_token_accuracy": 0.15591788440942764,
"num_tokens": 9692076.0,
"step": 5585
},
{
"entropy": 5.749252414703369,
"epoch": 0.4349348375802373,
"grad_norm": 1.1953125,
"learning_rate": 0.0004985636206573642,
"loss": 5.6574,
"mean_token_accuracy": 0.1567201316356659,
"num_tokens": 9700266.0,
"step": 5590
},
{
"entropy": 5.6495452404022215,
"epoch": 0.43532386695195485,
"grad_norm": 1.2265625,
"learning_rate": 0.0004985604922471575,
"loss": 5.5804,
"mean_token_accuracy": 0.15612577348947526,
"num_tokens": 9708643.0,
"step": 5595
},
{
"entropy": 5.722015714645385,
"epoch": 0.43571289632367244,
"grad_norm": 1.265625,
"learning_rate": 0.0004985573604447761,
"loss": 5.6391,
"mean_token_accuracy": 0.15270705446600913,
"num_tokens": 9717459.0,
"step": 5600
},
{
"entropy": 5.683666324615478,
"epoch": 0.43610192569539,
"grad_norm": 1.1875,
"learning_rate": 0.0004985542252502676,
"loss": 5.556,
"mean_token_accuracy": 0.15867960900068284,
"num_tokens": 9726251.0,
"step": 5605
},
{
"entropy": 5.572760534286499,
"epoch": 0.43649095506710756,
"grad_norm": 1.28125,
"learning_rate": 0.0004985510866636796,
"loss": 5.5414,
"mean_token_accuracy": 0.1563876047730446,
"num_tokens": 9735327.0,
"step": 5610
},
{
"entropy": 5.710997104644775,
"epoch": 0.43687998443882514,
"grad_norm": 1.2734375,
"learning_rate": 0.0004985479446850596,
"loss": 5.6939,
"mean_token_accuracy": 0.14747919738292695,
"num_tokens": 9743642.0,
"step": 5615
},
{
"entropy": 5.745656728744507,
"epoch": 0.4372690138105427,
"grad_norm": 1.2109375,
"learning_rate": 0.0004985447993144554,
"loss": 5.5575,
"mean_token_accuracy": 0.1631772004067898,
"num_tokens": 9752235.0,
"step": 5620
},
{
"entropy": 5.710186576843261,
"epoch": 0.43765804318226026,
"grad_norm": 1.21875,
"learning_rate": 0.0004985416505519147,
"loss": 5.6486,
"mean_token_accuracy": 0.14850047677755357,
"num_tokens": 9761012.0,
"step": 5625
},
{
"entropy": 5.664464712142944,
"epoch": 0.43804707255397785,
"grad_norm": 1.265625,
"learning_rate": 0.0004985384983974853,
"loss": 5.6143,
"mean_token_accuracy": 0.157231392711401,
"num_tokens": 9769810.0,
"step": 5630
},
{
"entropy": 5.700027561187744,
"epoch": 0.4384361019256954,
"grad_norm": 1.359375,
"learning_rate": 0.0004985353428512148,
"loss": 5.5438,
"mean_token_accuracy": 0.15658937096595765,
"num_tokens": 9778186.0,
"step": 5635
},
{
"entropy": 5.659540843963623,
"epoch": 0.43882513129741296,
"grad_norm": 1.2421875,
"learning_rate": 0.0004985321839131514,
"loss": 5.6229,
"mean_token_accuracy": 0.1591310068964958,
"num_tokens": 9786456.0,
"step": 5640
},
{
"entropy": 5.67944016456604,
"epoch": 0.4392141606691305,
"grad_norm": 1.2890625,
"learning_rate": 0.0004985290215833428,
"loss": 5.6157,
"mean_token_accuracy": 0.152775377035141,
"num_tokens": 9794613.0,
"step": 5645
},
{
"entropy": 5.751613807678223,
"epoch": 0.4396031900408481,
"grad_norm": 1.2734375,
"learning_rate": 0.0004985258558618372,
"loss": 5.5591,
"mean_token_accuracy": 0.15995817482471467,
"num_tokens": 9803388.0,
"step": 5650
},
{
"entropy": 5.620654487609864,
"epoch": 0.43999221941256567,
"grad_norm": 1.3359375,
"learning_rate": 0.0004985226867486825,
"loss": 5.6052,
"mean_token_accuracy": 0.16028854846954346,
"num_tokens": 9812450.0,
"step": 5655
},
{
"entropy": 5.6909432888031,
"epoch": 0.4403812487842832,
"grad_norm": 1.28125,
"learning_rate": 0.0004985195142439267,
"loss": 5.68,
"mean_token_accuracy": 0.15284086763858795,
"num_tokens": 9821377.0,
"step": 5660
},
{
"entropy": 5.7234701156616214,
"epoch": 0.4407702781560008,
"grad_norm": 1.4140625,
"learning_rate": 0.0004985163383476181,
"loss": 5.5778,
"mean_token_accuracy": 0.15479062795639037,
"num_tokens": 9829761.0,
"step": 5665
},
{
"entropy": 5.706999444961548,
"epoch": 0.44115930752771837,
"grad_norm": 1.359375,
"learning_rate": 0.0004985131590598048,
"loss": 5.6653,
"mean_token_accuracy": 0.14726217687129975,
"num_tokens": 9838921.0,
"step": 5670
},
{
"entropy": 5.646905469894409,
"epoch": 0.4415483368994359,
"grad_norm": 1.3125,
"learning_rate": 0.0004985099763805352,
"loss": 5.5007,
"mean_token_accuracy": 0.1659899964928627,
"num_tokens": 9846822.0,
"step": 5675
},
{
"entropy": 5.7456502437591555,
"epoch": 0.4419373662711535,
"grad_norm": 1.40625,
"learning_rate": 0.0004985067903098574,
"loss": 5.7295,
"mean_token_accuracy": 0.15629381388425828,
"num_tokens": 9855951.0,
"step": 5680
},
{
"entropy": 5.7572588443756105,
"epoch": 0.442326395642871,
"grad_norm": 1.28125,
"learning_rate": 0.0004985036008478197,
"loss": 5.6582,
"mean_token_accuracy": 0.15795501098036765,
"num_tokens": 9864875.0,
"step": 5685
},
{
"entropy": 5.738973951339721,
"epoch": 0.4427154250145886,
"grad_norm": 1.3359375,
"learning_rate": 0.0004985004079944707,
"loss": 5.6286,
"mean_token_accuracy": 0.1554968997836113,
"num_tokens": 9873785.0,
"step": 5690
},
{
"entropy": 5.714623546600341,
"epoch": 0.4431044543863062,
"grad_norm": 1.53125,
"learning_rate": 0.0004984972117498587,
"loss": 5.6291,
"mean_token_accuracy": 0.14927410408854486,
"num_tokens": 9883214.0,
"step": 5695
},
{
"entropy": 5.672214412689209,
"epoch": 0.4434934837580237,
"grad_norm": 1.546875,
"learning_rate": 0.0004984940121140323,
"loss": 5.6088,
"mean_token_accuracy": 0.15693314075469972,
"num_tokens": 9891346.0,
"step": 5700
},
{
"entropy": 5.637992000579834,
"epoch": 0.4438825131297413,
"grad_norm": 1.3515625,
"learning_rate": 0.00049849080908704,
"loss": 5.531,
"mean_token_accuracy": 0.15485147386789322,
"num_tokens": 9900646.0,
"step": 5705
},
{
"entropy": 5.71306209564209,
"epoch": 0.44427154250145884,
"grad_norm": 1.3515625,
"learning_rate": 0.0004984876026689302,
"loss": 5.6701,
"mean_token_accuracy": 0.15194630101323128,
"num_tokens": 9910851.0,
"step": 5710
},
{
"entropy": 5.784929990768433,
"epoch": 0.4446605718731764,
"grad_norm": 1.3359375,
"learning_rate": 0.000498484392859752,
"loss": 5.685,
"mean_token_accuracy": 0.15186055302619933,
"num_tokens": 9919919.0,
"step": 5715
},
{
"entropy": 5.767201137542725,
"epoch": 0.445049601244894,
"grad_norm": 1.40625,
"learning_rate": 0.0004984811796595538,
"loss": 5.6054,
"mean_token_accuracy": 0.15849266946315765,
"num_tokens": 9928831.0,
"step": 5720
},
{
"entropy": 5.662395191192627,
"epoch": 0.44543863061661154,
"grad_norm": 1.265625,
"learning_rate": 0.0004984779630683843,
"loss": 5.5571,
"mean_token_accuracy": 0.16009855270385742,
"num_tokens": 9936840.0,
"step": 5725
},
{
"entropy": 5.672893428802491,
"epoch": 0.44582765998832913,
"grad_norm": 1.203125,
"learning_rate": 0.0004984747430862924,
"loss": 5.647,
"mean_token_accuracy": 0.1535024493932724,
"num_tokens": 9945845.0,
"step": 5730
},
{
"entropy": 5.713797903060913,
"epoch": 0.44621668936004666,
"grad_norm": 1.328125,
"learning_rate": 0.0004984715197133271,
"loss": 5.6304,
"mean_token_accuracy": 0.1542983204126358,
"num_tokens": 9955176.0,
"step": 5735
},
{
"entropy": 5.6174845695495605,
"epoch": 0.44660571873176425,
"grad_norm": 1.34375,
"learning_rate": 0.0004984682929495371,
"loss": 5.5239,
"mean_token_accuracy": 0.16356904953718185,
"num_tokens": 9963182.0,
"step": 5740
},
{
"entropy": 5.703261232376098,
"epoch": 0.44699474810348183,
"grad_norm": 1.28125,
"learning_rate": 0.0004984650627949715,
"loss": 5.6546,
"mean_token_accuracy": 0.1533096879720688,
"num_tokens": 9971239.0,
"step": 5745
},
{
"entropy": 5.666077423095703,
"epoch": 0.44738377747519936,
"grad_norm": 1.25,
"learning_rate": 0.0004984618292496792,
"loss": 5.609,
"mean_token_accuracy": 0.1545335054397583,
"num_tokens": 9980285.0,
"step": 5750
},
{
"entropy": 5.693202304840088,
"epoch": 0.44777280684691695,
"grad_norm": 1.3515625,
"learning_rate": 0.0004984585923137093,
"loss": 5.6373,
"mean_token_accuracy": 0.15373781323432922,
"num_tokens": 9988922.0,
"step": 5755
},
{
"entropy": 5.741955804824829,
"epoch": 0.4481618362186345,
"grad_norm": 1.2734375,
"learning_rate": 0.000498455351987111,
"loss": 5.706,
"mean_token_accuracy": 0.14707285165786743,
"num_tokens": 9998538.0,
"step": 5760
},
{
"entropy": 5.788418865203857,
"epoch": 0.44855086559035207,
"grad_norm": 1.1953125,
"learning_rate": 0.0004984521082699333,
"loss": 5.6135,
"mean_token_accuracy": 0.15476753562688828,
"num_tokens": 10006646.0,
"step": 5765
},
{
"entropy": 5.661280012130737,
"epoch": 0.44893989496206965,
"grad_norm": 1.328125,
"learning_rate": 0.0004984488611622256,
"loss": 5.6027,
"mean_token_accuracy": 0.15705500692129135,
"num_tokens": 10016511.0,
"step": 5770
},
{
"entropy": 5.636836957931519,
"epoch": 0.4493289243337872,
"grad_norm": 1.2109375,
"learning_rate": 0.000498445610664037,
"loss": 5.4999,
"mean_token_accuracy": 0.1614679366350174,
"num_tokens": 10025223.0,
"step": 5775
},
{
"entropy": 5.649578857421875,
"epoch": 0.44971795370550477,
"grad_norm": 1.265625,
"learning_rate": 0.000498442356775417,
"loss": 5.5849,
"mean_token_accuracy": 0.15652598887681962,
"num_tokens": 10033678.0,
"step": 5780
},
{
"entropy": 5.66576738357544,
"epoch": 0.45010698307722236,
"grad_norm": 1.3203125,
"learning_rate": 0.0004984390994964148,
"loss": 5.4934,
"mean_token_accuracy": 0.15820170938968658,
"num_tokens": 10042028.0,
"step": 5785
},
{
"entropy": 5.672802352905274,
"epoch": 0.4504960124489399,
"grad_norm": 1.234375,
"learning_rate": 0.00049843583882708,
"loss": 5.6915,
"mean_token_accuracy": 0.1554969623684883,
"num_tokens": 10050758.0,
"step": 5790
},
{
"entropy": 5.641775178909302,
"epoch": 0.4508850418206575,
"grad_norm": 1.265625,
"learning_rate": 0.000498432574767462,
"loss": 5.4427,
"mean_token_accuracy": 0.16750287264585495,
"num_tokens": 10058688.0,
"step": 5795
},
{
"entropy": 5.644252109527588,
"epoch": 0.451274071192375,
"grad_norm": 1.234375,
"learning_rate": 0.0004984293073176103,
"loss": 5.5453,
"mean_token_accuracy": 0.1645413100719452,
"num_tokens": 10067302.0,
"step": 5800
},
{
"entropy": 5.5666821002960205,
"epoch": 0.4516631005640926,
"grad_norm": 1.21875,
"learning_rate": 0.0004984260364775744,
"loss": 5.5331,
"mean_token_accuracy": 0.15917521715164185,
"num_tokens": 10076424.0,
"step": 5805
},
{
"entropy": 5.724156904220581,
"epoch": 0.4520521299358102,
"grad_norm": 1.203125,
"learning_rate": 0.000498422762247404,
"loss": 5.6703,
"mean_token_accuracy": 0.14797478914260864,
"num_tokens": 10086307.0,
"step": 5810
},
{
"entropy": 5.696759462356567,
"epoch": 0.4524411593075277,
"grad_norm": 1.265625,
"learning_rate": 0.0004984194846271489,
"loss": 5.5395,
"mean_token_accuracy": 0.1550739273428917,
"num_tokens": 10094954.0,
"step": 5815
},
{
"entropy": 5.609128522872925,
"epoch": 0.4528301886792453,
"grad_norm": 1.2890625,
"learning_rate": 0.0004984162036168586,
"loss": 5.628,
"mean_token_accuracy": 0.15431981831789016,
"num_tokens": 10104401.0,
"step": 5820
},
{
"entropy": 5.638042354583741,
"epoch": 0.4532192180509628,
"grad_norm": 1.2890625,
"learning_rate": 0.0004984129192165831,
"loss": 5.625,
"mean_token_accuracy": 0.16010936647653579,
"num_tokens": 10112596.0,
"step": 5825
},
{
"entropy": 5.729681921005249,
"epoch": 0.4536082474226804,
"grad_norm": 1.3203125,
"learning_rate": 0.0004984096314263722,
"loss": 5.6688,
"mean_token_accuracy": 0.1558495357632637,
"num_tokens": 10121311.0,
"step": 5830
},
{
"entropy": 5.574494218826294,
"epoch": 0.453997276794398,
"grad_norm": 1.2421875,
"learning_rate": 0.0004984063402462757,
"loss": 5.468,
"mean_token_accuracy": 0.16243041604757308,
"num_tokens": 10130010.0,
"step": 5835
},
{
"entropy": 5.605675840377808,
"epoch": 0.45438630616611553,
"grad_norm": 1.2734375,
"learning_rate": 0.0004984030456763435,
"loss": 5.5912,
"mean_token_accuracy": 0.1590662494301796,
"num_tokens": 10138473.0,
"step": 5840
},
{
"entropy": 5.8279444694519045,
"epoch": 0.4547753355378331,
"grad_norm": 1.2890625,
"learning_rate": 0.0004983997477166257,
"loss": 5.651,
"mean_token_accuracy": 0.15664980113506316,
"num_tokens": 10146843.0,
"step": 5845
},
{
"entropy": 5.584151268005371,
"epoch": 0.45516436490955064,
"grad_norm": 1.2890625,
"learning_rate": 0.0004983964463671723,
"loss": 5.5015,
"mean_token_accuracy": 0.16206508874893188,
"num_tokens": 10155319.0,
"step": 5850
},
{
"entropy": 5.641605281829834,
"epoch": 0.45555339428126823,
"grad_norm": 1.375,
"learning_rate": 0.0004983931416280334,
"loss": 5.6853,
"mean_token_accuracy": 0.14777326062321663,
"num_tokens": 10164308.0,
"step": 5855
},
{
"entropy": 5.776509571075439,
"epoch": 0.4559424236529858,
"grad_norm": 1.296875,
"learning_rate": 0.0004983898334992591,
"loss": 5.7025,
"mean_token_accuracy": 0.1513962298631668,
"num_tokens": 10173173.0,
"step": 5860
},
{
"entropy": 5.691646337509155,
"epoch": 0.45633145302470335,
"grad_norm": 1.3984375,
"learning_rate": 0.0004983865219808998,
"loss": 5.6123,
"mean_token_accuracy": 0.15464477241039276,
"num_tokens": 10181376.0,
"step": 5865
},
{
"entropy": 5.724706697463989,
"epoch": 0.45672048239642093,
"grad_norm": 1.3046875,
"learning_rate": 0.0004983832070730055,
"loss": 5.6959,
"mean_token_accuracy": 0.1563864156603813,
"num_tokens": 10189701.0,
"step": 5870
},
{
"entropy": 5.710652160644531,
"epoch": 0.4571095117681385,
"grad_norm": 1.3828125,
"learning_rate": 0.0004983798887756265,
"loss": 5.5836,
"mean_token_accuracy": 0.1592947706580162,
"num_tokens": 10198721.0,
"step": 5875
},
{
"entropy": 5.672290182113647,
"epoch": 0.45749854113985605,
"grad_norm": 1.359375,
"learning_rate": 0.0004983765670888133,
"loss": 5.4761,
"mean_token_accuracy": 0.16269083470106124,
"num_tokens": 10206428.0,
"step": 5880
},
{
"entropy": 5.562413501739502,
"epoch": 0.45788757051157364,
"grad_norm": 1.296875,
"learning_rate": 0.0004983732420126163,
"loss": 5.6116,
"mean_token_accuracy": 0.15439745485782624,
"num_tokens": 10214941.0,
"step": 5885
},
{
"entropy": 5.67099027633667,
"epoch": 0.45827659988329117,
"grad_norm": 1.28125,
"learning_rate": 0.0004983699135470858,
"loss": 5.6174,
"mean_token_accuracy": 0.16162062734365462,
"num_tokens": 10223491.0,
"step": 5890
},
{
"entropy": 5.720311975479126,
"epoch": 0.45866562925500876,
"grad_norm": 1.3671875,
"learning_rate": 0.0004983665816922723,
"loss": 5.5396,
"mean_token_accuracy": 0.15808571577072145,
"num_tokens": 10231820.0,
"step": 5895
},
{
"entropy": 5.61470627784729,
"epoch": 0.45905465862672634,
"grad_norm": 1.265625,
"learning_rate": 0.0004983632464482267,
"loss": 5.536,
"mean_token_accuracy": 0.16108897626399993,
"num_tokens": 10239781.0,
"step": 5900
},
{
"entropy": 5.635873746871948,
"epoch": 0.4594436879984439,
"grad_norm": 1.328125,
"learning_rate": 0.0004983599078149991,
"loss": 5.5609,
"mean_token_accuracy": 0.16366778314113617,
"num_tokens": 10248093.0,
"step": 5905
},
{
"entropy": 5.69731912612915,
"epoch": 0.45983271737016146,
"grad_norm": 1.3046875,
"learning_rate": 0.0004983565657926405,
"loss": 5.5878,
"mean_token_accuracy": 0.1563509926199913,
"num_tokens": 10256370.0,
"step": 5910
},
{
"entropy": 5.619994688034057,
"epoch": 0.460221746741879,
"grad_norm": 1.25,
"learning_rate": 0.0004983532203812017,
"loss": 5.5593,
"mean_token_accuracy": 0.1569221630692482,
"num_tokens": 10264924.0,
"step": 5915
},
{
"entropy": 5.630159044265747,
"epoch": 0.4606107761135966,
"grad_norm": 1.3984375,
"learning_rate": 0.0004983498715807331,
"loss": 5.5033,
"mean_token_accuracy": 0.160485377907753,
"num_tokens": 10273488.0,
"step": 5920
},
{
"entropy": 5.672831010818482,
"epoch": 0.46099980548531416,
"grad_norm": 1.3125,
"learning_rate": 0.0004983465193912857,
"loss": 5.5644,
"mean_token_accuracy": 0.1572013482451439,
"num_tokens": 10281534.0,
"step": 5925
},
{
"entropy": 5.719956016540527,
"epoch": 0.4613888348570317,
"grad_norm": 1.3359375,
"learning_rate": 0.0004983431638129104,
"loss": 5.6069,
"mean_token_accuracy": 0.15728743895888328,
"num_tokens": 10289613.0,
"step": 5930
},
{
"entropy": 5.74925742149353,
"epoch": 0.4617778642287493,
"grad_norm": 1.375,
"learning_rate": 0.0004983398048456581,
"loss": 5.6668,
"mean_token_accuracy": 0.15779462456703186,
"num_tokens": 10299027.0,
"step": 5935
},
{
"entropy": 5.748346519470215,
"epoch": 0.4621668936004668,
"grad_norm": 1.296875,
"learning_rate": 0.0004983364424895796,
"loss": 5.6219,
"mean_token_accuracy": 0.15128087997436523,
"num_tokens": 10308727.0,
"step": 5940
},
{
"entropy": 5.716321325302124,
"epoch": 0.4625559229721844,
"grad_norm": 1.28125,
"learning_rate": 0.0004983330767447262,
"loss": 5.6393,
"mean_token_accuracy": 0.15531712099909784,
"num_tokens": 10318129.0,
"step": 5945
},
{
"entropy": 5.6274834156036375,
"epoch": 0.462944952343902,
"grad_norm": 1.3125,
"learning_rate": 0.0004983297076111489,
"loss": 5.494,
"mean_token_accuracy": 0.1664715051651001,
"num_tokens": 10326504.0,
"step": 5950
},
{
"entropy": 5.549721384048462,
"epoch": 0.4633339817156195,
"grad_norm": 1.34375,
"learning_rate": 0.0004983263350888987,
"loss": 5.5596,
"mean_token_accuracy": 0.16194145381450653,
"num_tokens": 10334585.0,
"step": 5955
},
{
"entropy": 5.6667708396911625,
"epoch": 0.4637230110873371,
"grad_norm": 1.3359375,
"learning_rate": 0.0004983229591780268,
"loss": 5.5495,
"mean_token_accuracy": 0.1623869687318802,
"num_tokens": 10343328.0,
"step": 5960
},
{
"entropy": 5.665079545974732,
"epoch": 0.46411204045905463,
"grad_norm": 1.265625,
"learning_rate": 0.0004983195798785844,
"loss": 5.5816,
"mean_token_accuracy": 0.15681301951408386,
"num_tokens": 10352289.0,
"step": 5965
},
{
"entropy": 5.705914878845215,
"epoch": 0.4645010698307722,
"grad_norm": 1.3984375,
"learning_rate": 0.0004983161971906228,
"loss": 5.6951,
"mean_token_accuracy": 0.14757215976715088,
"num_tokens": 10360847.0,
"step": 5970
},
{
"entropy": 5.761966180801392,
"epoch": 0.4648900992024898,
"grad_norm": 1.375,
"learning_rate": 0.0004983128111141935,
"loss": 5.6165,
"mean_token_accuracy": 0.1592383563518524,
"num_tokens": 10368911.0,
"step": 5975
},
{
"entropy": 5.662771272659302,
"epoch": 0.46527912857420733,
"grad_norm": 1.28125,
"learning_rate": 0.0004983094216493475,
"loss": 5.5573,
"mean_token_accuracy": 0.16440646350383759,
"num_tokens": 10377257.0,
"step": 5980
},
{
"entropy": 5.601254081726074,
"epoch": 0.4656681579459249,
"grad_norm": 1.3671875,
"learning_rate": 0.0004983060287961367,
"loss": 5.6192,
"mean_token_accuracy": 0.15791454762220383,
"num_tokens": 10386167.0,
"step": 5985
},
{
"entropy": 5.633698558807373,
"epoch": 0.4660571873176425,
"grad_norm": 1.203125,
"learning_rate": 0.0004983026325546123,
"loss": 5.5035,
"mean_token_accuracy": 0.1562596693634987,
"num_tokens": 10395802.0,
"step": 5990
},
{
"entropy": 5.68686261177063,
"epoch": 0.46644621668936004,
"grad_norm": 1.4921875,
"learning_rate": 0.0004982992329248257,
"loss": 5.6341,
"mean_token_accuracy": 0.15112658590078354,
"num_tokens": 10404836.0,
"step": 5995
},
{
"entropy": 5.764874982833862,
"epoch": 0.4668352460610776,
"grad_norm": 1.2734375,
"learning_rate": 0.0004982958299068289,
"loss": 5.623,
"mean_token_accuracy": 0.15592485070228576,
"num_tokens": 10413394.0,
"step": 6000
},
{
"epoch": 0.4668352460610776,
"eval_entropy": 5.559448653855384,
"eval_loss": 5.644718647003174,
"eval_mean_token_accuracy": 0.16176202593584607,
"eval_num_tokens": 10413394.0,
"eval_runtime": 28.4474,
"eval_samples_per_second": 1460.874,
"eval_steps_per_second": 182.618,
"step": 6000
},
{
"entropy": 5.599856758117676,
"epoch": 0.46722427543279516,
"grad_norm": 1.390625,
"learning_rate": 0.000498292423500673,
"loss": 5.5164,
"mean_token_accuracy": 0.16224406957626342,
"num_tokens": 10421916.0,
"step": 6005
},
{
"entropy": 5.646891021728516,
"epoch": 0.46761330480451274,
"grad_norm": 1.3203125,
"learning_rate": 0.0004982890137064102,
"loss": 5.582,
"mean_token_accuracy": 0.15907793641090393,
"num_tokens": 10430703.0,
"step": 6010
},
{
"entropy": 5.743211698532105,
"epoch": 0.46800233417623033,
"grad_norm": 1.3203125,
"learning_rate": 0.000498285600524092,
"loss": 5.6429,
"mean_token_accuracy": 0.1553155668079853,
"num_tokens": 10439523.0,
"step": 6015
},
{
"entropy": 5.683924865722656,
"epoch": 0.46839136354794786,
"grad_norm": 1.5390625,
"learning_rate": 0.0004982821839537701,
"loss": 5.5435,
"mean_token_accuracy": 0.1604115754365921,
"num_tokens": 10447777.0,
"step": 6020
},
{
"entropy": 5.6141105651855465,
"epoch": 0.46878039291966545,
"grad_norm": 1.3203125,
"learning_rate": 0.0004982787639954966,
"loss": 5.5246,
"mean_token_accuracy": 0.1618039131164551,
"num_tokens": 10456105.0,
"step": 6025
},
{
"entropy": 5.640907192230225,
"epoch": 0.469169422291383,
"grad_norm": 1.3359375,
"learning_rate": 0.0004982753406493232,
"loss": 5.5704,
"mean_token_accuracy": 0.159077899903059,
"num_tokens": 10464590.0,
"step": 6030
},
{
"entropy": 5.687194967269898,
"epoch": 0.46955845166310056,
"grad_norm": 1.5078125,
"learning_rate": 0.0004982719139153018,
"loss": 5.5938,
"mean_token_accuracy": 0.1523827828466892,
"num_tokens": 10472702.0,
"step": 6035
},
{
"entropy": 5.57777214050293,
"epoch": 0.46994748103481815,
"grad_norm": 1.3125,
"learning_rate": 0.0004982684837934845,
"loss": 5.5136,
"mean_token_accuracy": 0.1578451305627823,
"num_tokens": 10480682.0,
"step": 6040
},
{
"entropy": 5.688637924194336,
"epoch": 0.4703365104065357,
"grad_norm": 1.3359375,
"learning_rate": 0.0004982650502839234,
"loss": 5.6146,
"mean_token_accuracy": 0.15147078782320023,
"num_tokens": 10488873.0,
"step": 6045
},
{
"entropy": 5.66636176109314,
"epoch": 0.47072553977825327,
"grad_norm": 1.2265625,
"learning_rate": 0.0004982616133866705,
"loss": 5.5789,
"mean_token_accuracy": 0.16179264783859254,
"num_tokens": 10497077.0,
"step": 6050
},
{
"entropy": 5.681707859039307,
"epoch": 0.4711145691499708,
"grad_norm": 1.2734375,
"learning_rate": 0.0004982581731017779,
"loss": 5.5575,
"mean_token_accuracy": 0.15877941399812698,
"num_tokens": 10506072.0,
"step": 6055
},
{
"entropy": 5.718855381011963,
"epoch": 0.4715035985216884,
"grad_norm": 1.34375,
"learning_rate": 0.0004982547294292979,
"loss": 5.6207,
"mean_token_accuracy": 0.15153614282608033,
"num_tokens": 10515175.0,
"step": 6060
},
{
"entropy": 5.605925559997559,
"epoch": 0.47189262789340597,
"grad_norm": 1.3515625,
"learning_rate": 0.0004982512823692828,
"loss": 5.508,
"mean_token_accuracy": 0.15940281450748445,
"num_tokens": 10524800.0,
"step": 6065
},
{
"entropy": 5.652846336364746,
"epoch": 0.4722816572651235,
"grad_norm": 1.40625,
"learning_rate": 0.0004982478319217848,
"loss": 5.7054,
"mean_token_accuracy": 0.14966497719287872,
"num_tokens": 10534608.0,
"step": 6070
},
{
"entropy": 5.745728874206543,
"epoch": 0.4726706866368411,
"grad_norm": 1.2421875,
"learning_rate": 0.0004982443780868562,
"loss": 5.5948,
"mean_token_accuracy": 0.15881826132535934,
"num_tokens": 10542992.0,
"step": 6075
},
{
"entropy": 5.662547779083252,
"epoch": 0.4730597160085587,
"grad_norm": 1.3515625,
"learning_rate": 0.0004982409208645496,
"loss": 5.6235,
"mean_token_accuracy": 0.15793444067239762,
"num_tokens": 10551682.0,
"step": 6080
},
{
"entropy": 5.607632875442505,
"epoch": 0.4734487453802762,
"grad_norm": 1.25,
"learning_rate": 0.0004982374602549173,
"loss": 5.55,
"mean_token_accuracy": 0.16171603202819823,
"num_tokens": 10560556.0,
"step": 6085
},
{
"entropy": 5.732918691635132,
"epoch": 0.4738377747519938,
"grad_norm": 1.359375,
"learning_rate": 0.0004982339962580119,
"loss": 5.7323,
"mean_token_accuracy": 0.14882105439901352,
"num_tokens": 10569361.0,
"step": 6090
},
{
"entropy": 5.668002557754517,
"epoch": 0.4742268041237113,
"grad_norm": 1.2421875,
"learning_rate": 0.0004982305288738859,
"loss": 5.5426,
"mean_token_accuracy": 0.16289373785257338,
"num_tokens": 10577997.0,
"step": 6095
},
{
"entropy": 5.701682186126709,
"epoch": 0.4746158334954289,
"grad_norm": 1.265625,
"learning_rate": 0.0004982270581025919,
"loss": 5.6444,
"mean_token_accuracy": 0.1589728429913521,
"num_tokens": 10587005.0,
"step": 6100
},
{
"entropy": 5.65606369972229,
"epoch": 0.4750048628671465,
"grad_norm": 1.375,
"learning_rate": 0.0004982235839441826,
"loss": 5.5047,
"mean_token_accuracy": 0.17128463685512543,
"num_tokens": 10595196.0,
"step": 6105
},
{
"entropy": 5.691293668746948,
"epoch": 0.475393892238864,
"grad_norm": 1.359375,
"learning_rate": 0.0004982201063987108,
"loss": 5.6487,
"mean_token_accuracy": 0.14815771952271461,
"num_tokens": 10603796.0,
"step": 6110
},
{
"entropy": 5.697172737121582,
"epoch": 0.4757829216105816,
"grad_norm": 1.3359375,
"learning_rate": 0.0004982166254662292,
"loss": 5.6324,
"mean_token_accuracy": 0.15333464592695237,
"num_tokens": 10612895.0,
"step": 6115
},
{
"entropy": 5.72194128036499,
"epoch": 0.47617195098229914,
"grad_norm": 1.3359375,
"learning_rate": 0.0004982131411467904,
"loss": 5.6968,
"mean_token_accuracy": 0.1535396084189415,
"num_tokens": 10621310.0,
"step": 6120
},
{
"entropy": 5.626172780990601,
"epoch": 0.4765609803540167,
"grad_norm": 1.3828125,
"learning_rate": 0.0004982096534404476,
"loss": 5.5128,
"mean_token_accuracy": 0.16079071760177613,
"num_tokens": 10629598.0,
"step": 6125
},
{
"entropy": 5.72747836112976,
"epoch": 0.4769500097257343,
"grad_norm": 1.359375,
"learning_rate": 0.0004982061623472535,
"loss": 5.5331,
"mean_token_accuracy": 0.15735719203948975,
"num_tokens": 10638218.0,
"step": 6130
},
{
"entropy": 5.687162113189697,
"epoch": 0.47733903909745184,
"grad_norm": 1.34375,
"learning_rate": 0.0004982026678672612,
"loss": 5.6234,
"mean_token_accuracy": 0.1520923927426338,
"num_tokens": 10647215.0,
"step": 6135
},
{
"entropy": 5.688553428649902,
"epoch": 0.47772806846916943,
"grad_norm": 1.28125,
"learning_rate": 0.0004981991700005238,
"loss": 5.7478,
"mean_token_accuracy": 0.14633624255657196,
"num_tokens": 10655755.0,
"step": 6140
},
{
"entropy": 5.761098909378052,
"epoch": 0.47811709784088696,
"grad_norm": 1.359375,
"learning_rate": 0.000498195668747094,
"loss": 5.6224,
"mean_token_accuracy": 0.1543148413300514,
"num_tokens": 10663969.0,
"step": 6145
},
{
"entropy": 5.714224195480346,
"epoch": 0.47850612721260455,
"grad_norm": 1.3046875,
"learning_rate": 0.0004981921641070254,
"loss": 5.493,
"mean_token_accuracy": 0.16152573823928834,
"num_tokens": 10672048.0,
"step": 6150
},
{
"entropy": 5.627324819564819,
"epoch": 0.47889515658432213,
"grad_norm": 1.2421875,
"learning_rate": 0.0004981886560803708,
"loss": 5.6021,
"mean_token_accuracy": 0.15429380685091018,
"num_tokens": 10681802.0,
"step": 6155
},
{
"entropy": 5.68464241027832,
"epoch": 0.47928418595603967,
"grad_norm": 1.453125,
"learning_rate": 0.0004981851446671838,
"loss": 5.6538,
"mean_token_accuracy": 0.14972650855779648,
"num_tokens": 10690618.0,
"step": 6160
},
{
"entropy": 5.689493703842163,
"epoch": 0.47967321532775725,
"grad_norm": 1.3515625,
"learning_rate": 0.0004981816298675173,
"loss": 5.5716,
"mean_token_accuracy": 0.15978924930095673,
"num_tokens": 10699080.0,
"step": 6165
},
{
"entropy": 5.633434629440307,
"epoch": 0.4800622446994748,
"grad_norm": 1.203125,
"learning_rate": 0.0004981781116814248,
"loss": 5.5734,
"mean_token_accuracy": 0.15396522134542465,
"num_tokens": 10708939.0,
"step": 6170
},
{
"entropy": 5.723195362091064,
"epoch": 0.48045127407119237,
"grad_norm": 1.3515625,
"learning_rate": 0.0004981745901089596,
"loss": 5.4795,
"mean_token_accuracy": 0.15833590775728226,
"num_tokens": 10717356.0,
"step": 6175
},
{
"entropy": 5.667737150192261,
"epoch": 0.48084030344290996,
"grad_norm": 1.3828125,
"learning_rate": 0.0004981710651501753,
"loss": 5.568,
"mean_token_accuracy": 0.15206077918410302,
"num_tokens": 10725616.0,
"step": 6180
},
{
"entropy": 5.615799808502198,
"epoch": 0.4812293328146275,
"grad_norm": 1.3671875,
"learning_rate": 0.0004981675368051254,
"loss": 5.5753,
"mean_token_accuracy": 0.1609847441315651,
"num_tokens": 10734337.0,
"step": 6185
},
{
"entropy": 5.684136199951172,
"epoch": 0.4816183621863451,
"grad_norm": 1.3125,
"learning_rate": 0.0004981640050738633,
"loss": 5.6006,
"mean_token_accuracy": 0.15555059164762497,
"num_tokens": 10742720.0,
"step": 6190
},
{
"entropy": 5.7327721118927,
"epoch": 0.48200739155806266,
"grad_norm": 1.4453125,
"learning_rate": 0.0004981604699564426,
"loss": 5.6446,
"mean_token_accuracy": 0.14744160026311875,
"num_tokens": 10751809.0,
"step": 6195
},
{
"entropy": 5.6392982006073,
"epoch": 0.4823964209297802,
"grad_norm": 1.234375,
"learning_rate": 0.0004981569314529169,
"loss": 5.5001,
"mean_token_accuracy": 0.15971516519784928,
"num_tokens": 10759879.0,
"step": 6200
},
{
"entropy": 5.5582897663116455,
"epoch": 0.4827854503014978,
"grad_norm": 1.390625,
"learning_rate": 0.00049815338956334,
"loss": 5.5642,
"mean_token_accuracy": 0.15791321396827698,
"num_tokens": 10769309.0,
"step": 6205
},
{
"entropy": 5.654816770553589,
"epoch": 0.4831744796732153,
"grad_norm": 1.3515625,
"learning_rate": 0.0004981498442877656,
"loss": 5.4911,
"mean_token_accuracy": 0.16367221474647523,
"num_tokens": 10777676.0,
"step": 6210
},
{
"entropy": 5.650180530548096,
"epoch": 0.4835635090449329,
"grad_norm": 1.4140625,
"learning_rate": 0.0004981462956262474,
"loss": 5.708,
"mean_token_accuracy": 0.15045704692602158,
"num_tokens": 10786278.0,
"step": 6215
},
{
"entropy": 5.742503547668457,
"epoch": 0.4839525384166505,
"grad_norm": 1.21875,
"learning_rate": 0.0004981427435788396,
"loss": 5.6291,
"mean_token_accuracy": 0.15221197307109832,
"num_tokens": 10795323.0,
"step": 6220
},
{
"entropy": 5.7140473365783695,
"epoch": 0.484341567788368,
"grad_norm": 1.3515625,
"learning_rate": 0.0004981391881455957,
"loss": 5.5261,
"mean_token_accuracy": 0.15894377529621123,
"num_tokens": 10803978.0,
"step": 6225
},
{
"entropy": 5.671517276763916,
"epoch": 0.4847305971600856,
"grad_norm": 1.3046875,
"learning_rate": 0.0004981356293265696,
"loss": 5.6268,
"mean_token_accuracy": 0.15697673261165618,
"num_tokens": 10812488.0,
"step": 6230
},
{
"entropy": 5.6791136264801025,
"epoch": 0.4851196265318031,
"grad_norm": 1.3046875,
"learning_rate": 0.0004981320671218157,
"loss": 5.5952,
"mean_token_accuracy": 0.1559137649834156,
"num_tokens": 10821076.0,
"step": 6235
},
{
"entropy": 5.59618444442749,
"epoch": 0.4855086559035207,
"grad_norm": 1.3046875,
"learning_rate": 0.0004981285015313877,
"loss": 5.424,
"mean_token_accuracy": 0.17230252772569657,
"num_tokens": 10828634.0,
"step": 6240
},
{
"entropy": 5.599424219131469,
"epoch": 0.4858976852752383,
"grad_norm": 1.5546875,
"learning_rate": 0.0004981249325553399,
"loss": 5.5609,
"mean_token_accuracy": 0.15971072167158126,
"num_tokens": 10837481.0,
"step": 6245
},
{
"entropy": 5.683236837387085,
"epoch": 0.48628671464695583,
"grad_norm": 1.328125,
"learning_rate": 0.0004981213601937264,
"loss": 5.6434,
"mean_token_accuracy": 0.15290929526090621,
"num_tokens": 10846729.0,
"step": 6250
},
{
"entropy": 5.744317626953125,
"epoch": 0.4866757440186734,
"grad_norm": 1.390625,
"learning_rate": 0.0004981177844466013,
"loss": 5.616,
"mean_token_accuracy": 0.15153464525938035,
"num_tokens": 10855344.0,
"step": 6255
},
{
"entropy": 5.702961158752442,
"epoch": 0.48706477339039095,
"grad_norm": 1.5,
"learning_rate": 0.0004981142053140192,
"loss": 5.6496,
"mean_token_accuracy": 0.15403816998004913,
"num_tokens": 10863934.0,
"step": 6260
},
{
"entropy": 5.659459257125855,
"epoch": 0.48745380276210853,
"grad_norm": 1.3984375,
"learning_rate": 0.0004981106227960339,
"loss": 5.6108,
"mean_token_accuracy": 0.15299373418092727,
"num_tokens": 10872031.0,
"step": 6265
},
{
"entropy": 5.730347967147827,
"epoch": 0.4878428321338261,
"grad_norm": 1.3203125,
"learning_rate": 0.0004981070368927001,
"loss": 5.6902,
"mean_token_accuracy": 0.15396978259086608,
"num_tokens": 10881385.0,
"step": 6270
},
{
"entropy": 5.695536041259766,
"epoch": 0.48823186150554365,
"grad_norm": 1.421875,
"learning_rate": 0.000498103447604072,
"loss": 5.5999,
"mean_token_accuracy": 0.16020937711000444,
"num_tokens": 10890483.0,
"step": 6275
},
{
"entropy": 5.704893159866333,
"epoch": 0.48862089087726124,
"grad_norm": 1.3515625,
"learning_rate": 0.0004980998549302044,
"loss": 5.6332,
"mean_token_accuracy": 0.16258720457553863,
"num_tokens": 10898338.0,
"step": 6280
},
{
"entropy": 5.659977054595947,
"epoch": 0.4890099202489788,
"grad_norm": 1.4453125,
"learning_rate": 0.0004980962588711516,
"loss": 5.6093,
"mean_token_accuracy": 0.1559084713459015,
"num_tokens": 10908087.0,
"step": 6285
},
{
"entropy": 5.747127103805542,
"epoch": 0.48939894962069636,
"grad_norm": 1.328125,
"learning_rate": 0.000498092659426968,
"loss": 5.6047,
"mean_token_accuracy": 0.14977360516786575,
"num_tokens": 10916704.0,
"step": 6290
},
{
"entropy": 5.733406352996826,
"epoch": 0.48978797899241394,
"grad_norm": 1.6328125,
"learning_rate": 0.0004980890565977085,
"loss": 5.5708,
"mean_token_accuracy": 0.15654229521751403,
"num_tokens": 10925186.0,
"step": 6295
},
{
"entropy": 5.670359992980957,
"epoch": 0.4901770083641315,
"grad_norm": 1.3671875,
"learning_rate": 0.0004980854503834276,
"loss": 5.6237,
"mean_token_accuracy": 0.15868111401796342,
"num_tokens": 10934247.0,
"step": 6300
},
{
"entropy": 5.730028200149536,
"epoch": 0.49056603773584906,
"grad_norm": 1.3125,
"learning_rate": 0.0004980818407841802,
"loss": 5.6388,
"mean_token_accuracy": 0.15077922344207764,
"num_tokens": 10944260.0,
"step": 6305
},
{
"entropy": 5.684900522232056,
"epoch": 0.49095506710756665,
"grad_norm": 1.3046875,
"learning_rate": 0.0004980782278000207,
"loss": 5.6024,
"mean_token_accuracy": 0.15438488870859146,
"num_tokens": 10954065.0,
"step": 6310
},
{
"entropy": 5.69667501449585,
"epoch": 0.4913440964792842,
"grad_norm": 1.296875,
"learning_rate": 0.0004980746114310043,
"loss": 5.6701,
"mean_token_accuracy": 0.15619819164276122,
"num_tokens": 10962918.0,
"step": 6315
},
{
"entropy": 5.599684524536133,
"epoch": 0.49173312585100176,
"grad_norm": 1.328125,
"learning_rate": 0.0004980709916771858,
"loss": 5.5242,
"mean_token_accuracy": 0.1607260599732399,
"num_tokens": 10972356.0,
"step": 6320
},
{
"entropy": 5.636107015609741,
"epoch": 0.4921221552227193,
"grad_norm": 1.453125,
"learning_rate": 0.0004980673685386198,
"loss": 5.5757,
"mean_token_accuracy": 0.15274077281355858,
"num_tokens": 10980984.0,
"step": 6325
},
{
"entropy": 5.631813192367554,
"epoch": 0.4925111845944369,
"grad_norm": 1.3046875,
"learning_rate": 0.0004980637420153618,
"loss": 5.5111,
"mean_token_accuracy": 0.15772687792778015,
"num_tokens": 10989809.0,
"step": 6330
},
{
"entropy": 5.665310287475586,
"epoch": 0.49290021396615447,
"grad_norm": 1.3359375,
"learning_rate": 0.0004980601121074664,
"loss": 5.5641,
"mean_token_accuracy": 0.1530244916677475,
"num_tokens": 10998824.0,
"step": 6335
},
{
"entropy": 5.738433790206909,
"epoch": 0.493289243337872,
"grad_norm": 1.3203125,
"learning_rate": 0.0004980564788149889,
"loss": 5.6617,
"mean_token_accuracy": 0.15882298350334167,
"num_tokens": 11006755.0,
"step": 6340
},
{
"entropy": 5.645573234558105,
"epoch": 0.4936782727095896,
"grad_norm": 1.3203125,
"learning_rate": 0.0004980528421379844,
"loss": 5.5346,
"mean_token_accuracy": 0.16477559357881547,
"num_tokens": 11015337.0,
"step": 6345
},
{
"entropy": 5.5610456466674805,
"epoch": 0.4940673020813071,
"grad_norm": 1.3671875,
"learning_rate": 0.000498049202076508,
"loss": 5.4638,
"mean_token_accuracy": 0.16363136917352678,
"num_tokens": 11024186.0,
"step": 6350
},
{
"entropy": 5.689836311340332,
"epoch": 0.4944563314530247,
"grad_norm": 1.2890625,
"learning_rate": 0.0004980455586306149,
"loss": 5.586,
"mean_token_accuracy": 0.16271042674779893,
"num_tokens": 11032504.0,
"step": 6355
},
{
"entropy": 5.724547481536865,
"epoch": 0.4948453608247423,
"grad_norm": 1.4375,
"learning_rate": 0.0004980419118003605,
"loss": 5.5451,
"mean_token_accuracy": 0.1571075886487961,
"num_tokens": 11042252.0,
"step": 6360
},
{
"entropy": 5.6988561153411865,
"epoch": 0.4952343901964598,
"grad_norm": 1.34375,
"learning_rate": 0.0004980382615858001,
"loss": 5.6027,
"mean_token_accuracy": 0.15843562185764312,
"num_tokens": 11051361.0,
"step": 6365
},
{
"entropy": 5.635060262680054,
"epoch": 0.4956234195681774,
"grad_norm": 1.328125,
"learning_rate": 0.0004980346079869892,
"loss": 5.5843,
"mean_token_accuracy": 0.1544535830616951,
"num_tokens": 11060822.0,
"step": 6370
},
{
"entropy": 5.700907468795776,
"epoch": 0.49601244893989493,
"grad_norm": 1.53125,
"learning_rate": 0.000498030951003983,
"loss": 5.6455,
"mean_token_accuracy": 0.1547166809439659,
"num_tokens": 11069528.0,
"step": 6375
},
{
"entropy": 5.679643297195435,
"epoch": 0.4964014783116125,
"grad_norm": 1.265625,
"learning_rate": 0.0004980272906368371,
"loss": 5.4827,
"mean_token_accuracy": 0.16679324954748154,
"num_tokens": 11077917.0,
"step": 6380
},
{
"entropy": 5.59128475189209,
"epoch": 0.4967905076833301,
"grad_norm": 1.203125,
"learning_rate": 0.0004980236268856071,
"loss": 5.5546,
"mean_token_accuracy": 0.1547350376844406,
"num_tokens": 11085938.0,
"step": 6385
},
{
"entropy": 5.609622240066528,
"epoch": 0.49717953705504764,
"grad_norm": 1.234375,
"learning_rate": 0.0004980199597503487,
"loss": 5.6147,
"mean_token_accuracy": 0.15192969441413878,
"num_tokens": 11094898.0,
"step": 6390
},
{
"entropy": 5.693767881393432,
"epoch": 0.4975685664267652,
"grad_norm": 1.390625,
"learning_rate": 0.0004980162892311171,
"loss": 5.4737,
"mean_token_accuracy": 0.16259513646364213,
"num_tokens": 11103038.0,
"step": 6395
},
{
"entropy": 5.697658061981201,
"epoch": 0.4979575957984828,
"grad_norm": 1.3515625,
"learning_rate": 0.0004980126153279684,
"loss": 5.5723,
"mean_token_accuracy": 0.15827525407075882,
"num_tokens": 11111965.0,
"step": 6400
},
{
"entropy": 5.605089139938355,
"epoch": 0.49834662517020034,
"grad_norm": 1.25,
"learning_rate": 0.0004980089380409583,
"loss": 5.5947,
"mean_token_accuracy": 0.15535541325807573,
"num_tokens": 11121482.0,
"step": 6405
},
{
"entropy": 5.672932577133179,
"epoch": 0.4987356545419179,
"grad_norm": 1.3046875,
"learning_rate": 0.0004980052573701424,
"loss": 5.5853,
"mean_token_accuracy": 0.15607178062200547,
"num_tokens": 11129742.0,
"step": 6410
},
{
"entropy": 5.675456905364991,
"epoch": 0.49912468391363546,
"grad_norm": 1.1953125,
"learning_rate": 0.0004980015733155767,
"loss": 5.5873,
"mean_token_accuracy": 0.15734535753726958,
"num_tokens": 11138388.0,
"step": 6415
},
{
"entropy": 5.520279121398926,
"epoch": 0.49951371328535304,
"grad_norm": 1.375,
"learning_rate": 0.0004979978858773171,
"loss": 5.3289,
"mean_token_accuracy": 0.17224133461713792,
"num_tokens": 11146967.0,
"step": 6420
},
{
"entropy": 5.645619964599609,
"epoch": 0.49990274265707063,
"grad_norm": 1.3125,
"learning_rate": 0.0004979941950554195,
"loss": 5.6735,
"mean_token_accuracy": 0.15287676751613616,
"num_tokens": 11155888.0,
"step": 6425
},
{
"entropy": 5.717220449447632,
"epoch": 0.5002917720287882,
"grad_norm": 1.3359375,
"learning_rate": 0.0004979905008499399,
"loss": 5.5328,
"mean_token_accuracy": 0.15672664046287538,
"num_tokens": 11164358.0,
"step": 6430
},
{
"entropy": 5.59271731376648,
"epoch": 0.5006808014005057,
"grad_norm": 1.265625,
"learning_rate": 0.0004979868032609344,
"loss": 5.3922,
"mean_token_accuracy": 0.17046331167221068,
"num_tokens": 11172416.0,
"step": 6435
},
{
"entropy": 5.661533260345459,
"epoch": 0.5010698307722233,
"grad_norm": 1.3203125,
"learning_rate": 0.0004979831022884591,
"loss": 5.5715,
"mean_token_accuracy": 0.15518891513347627,
"num_tokens": 11181159.0,
"step": 6440
},
{
"entropy": 5.6787347316741945,
"epoch": 0.5014588601439409,
"grad_norm": 1.375,
"learning_rate": 0.0004979793979325701,
"loss": 5.6135,
"mean_token_accuracy": 0.15603988468647004,
"num_tokens": 11189725.0,
"step": 6445
},
{
"entropy": 5.678800773620606,
"epoch": 0.5018478895156584,
"grad_norm": 1.34375,
"learning_rate": 0.0004979756901933236,
"loss": 5.5999,
"mean_token_accuracy": 0.15560702234506607,
"num_tokens": 11197795.0,
"step": 6450
},
{
"entropy": 5.674616813659668,
"epoch": 0.502236918887376,
"grad_norm": 1.296875,
"learning_rate": 0.000497971979070776,
"loss": 5.5438,
"mean_token_accuracy": 0.16058342829346656,
"num_tokens": 11205352.0,
"step": 6455
},
{
"entropy": 5.674797487258911,
"epoch": 0.5026259482590936,
"grad_norm": 1.375,
"learning_rate": 0.0004979682645649834,
"loss": 5.6084,
"mean_token_accuracy": 0.1550489269196987,
"num_tokens": 11214142.0,
"step": 6460
},
{
"entropy": 5.61780571937561,
"epoch": 0.5030149776308112,
"grad_norm": 1.234375,
"learning_rate": 0.0004979645466760025,
"loss": 5.4781,
"mean_token_accuracy": 0.16321058422327042,
"num_tokens": 11222565.0,
"step": 6465
},
{
"entropy": 5.588930559158325,
"epoch": 0.5034040070025287,
"grad_norm": 1.2421875,
"learning_rate": 0.0004979608254038892,
"loss": 5.5811,
"mean_token_accuracy": 0.15546856224536895,
"num_tokens": 11231829.0,
"step": 6470
},
{
"entropy": 5.624651479721069,
"epoch": 0.5037930363742462,
"grad_norm": 1.984375,
"learning_rate": 0.0004979571007487003,
"loss": 5.5923,
"mean_token_accuracy": 0.15631648153066635,
"num_tokens": 11240302.0,
"step": 6475
},
{
"entropy": 5.791405725479126,
"epoch": 0.5041820657459638,
"grad_norm": 1.4609375,
"learning_rate": 0.0004979533727104924,
"loss": 5.6838,
"mean_token_accuracy": 0.1518307164311409,
"num_tokens": 11248610.0,
"step": 6480
},
{
"entropy": 5.69047703742981,
"epoch": 0.5045710951176814,
"grad_norm": 1.28125,
"learning_rate": 0.000497949641289322,
"loss": 5.5964,
"mean_token_accuracy": 0.15891549587249756,
"num_tokens": 11257441.0,
"step": 6485
},
{
"entropy": 5.549295377731323,
"epoch": 0.504960124489399,
"grad_norm": 1.328125,
"learning_rate": 0.0004979459064852456,
"loss": 5.4659,
"mean_token_accuracy": 0.15760838985443115,
"num_tokens": 11265915.0,
"step": 6490
},
{
"entropy": 5.648113203048706,
"epoch": 0.5053491538611166,
"grad_norm": 1.3359375,
"learning_rate": 0.0004979421682983197,
"loss": 5.5975,
"mean_token_accuracy": 0.15131368562579156,
"num_tokens": 11274165.0,
"step": 6495
},
{
"entropy": 5.706796312332154,
"epoch": 0.505738183232834,
"grad_norm": 1.3125,
"learning_rate": 0.0004979384267286015,
"loss": 5.5461,
"mean_token_accuracy": 0.16159669756889344,
"num_tokens": 11283316.0,
"step": 6500
},
{
"entropy": 5.690013027191162,
"epoch": 0.5061272126045516,
"grad_norm": 1.359375,
"learning_rate": 0.0004979346817761475,
"loss": 5.6692,
"mean_token_accuracy": 0.15432394593954085,
"num_tokens": 11292289.0,
"step": 6505
},
{
"entropy": 5.678025627136231,
"epoch": 0.5065162419762692,
"grad_norm": 1.2265625,
"learning_rate": 0.0004979309334410144,
"loss": 5.5717,
"mean_token_accuracy": 0.15459866374731063,
"num_tokens": 11301192.0,
"step": 6510
},
{
"entropy": 5.6778205871582035,
"epoch": 0.5069052713479868,
"grad_norm": 1.3359375,
"learning_rate": 0.0004979271817232594,
"loss": 5.5882,
"mean_token_accuracy": 0.1566533237695694,
"num_tokens": 11309906.0,
"step": 6515
},
{
"entropy": 5.721369457244873,
"epoch": 0.5072943007197044,
"grad_norm": 1.3515625,
"learning_rate": 0.0004979234266229391,
"loss": 5.5822,
"mean_token_accuracy": 0.15548757910728456,
"num_tokens": 11318227.0,
"step": 6520
},
{
"entropy": 5.576367330551148,
"epoch": 0.5076833300914219,
"grad_norm": 1.3203125,
"learning_rate": 0.0004979196681401106,
"loss": 5.5998,
"mean_token_accuracy": 0.1588978312909603,
"num_tokens": 11327524.0,
"step": 6525
},
{
"entropy": 5.630653429031372,
"epoch": 0.5080723594631394,
"grad_norm": 1.3671875,
"learning_rate": 0.000497915906274831,
"loss": 5.5438,
"mean_token_accuracy": 0.161087529361248,
"num_tokens": 11336214.0,
"step": 6530
},
{
"entropy": 5.631156921386719,
"epoch": 0.508461388834857,
"grad_norm": 1.359375,
"learning_rate": 0.0004979121410271574,
"loss": 5.3896,
"mean_token_accuracy": 0.16864126175642014,
"num_tokens": 11344551.0,
"step": 6535
},
{
"entropy": 5.530792999267578,
"epoch": 0.5088504182065746,
"grad_norm": 1.3125,
"learning_rate": 0.0004979083723971468,
"loss": 5.46,
"mean_token_accuracy": 0.16636885404586793,
"num_tokens": 11353180.0,
"step": 6540
},
{
"entropy": 5.495325994491577,
"epoch": 0.5092394475782922,
"grad_norm": 1.21875,
"learning_rate": 0.0004979046003848564,
"loss": 5.4368,
"mean_token_accuracy": 0.16852892339229583,
"num_tokens": 11361343.0,
"step": 6545
},
{
"entropy": 5.625004386901855,
"epoch": 0.5096284769500097,
"grad_norm": 1.3125,
"learning_rate": 0.0004979008249903435,
"loss": 5.5466,
"mean_token_accuracy": 0.16107016056776047,
"num_tokens": 11369915.0,
"step": 6550
},
{
"entropy": 5.6815948486328125,
"epoch": 0.5100175063217273,
"grad_norm": 1.2265625,
"learning_rate": 0.0004978970462136655,
"loss": 5.564,
"mean_token_accuracy": 0.15614417046308518,
"num_tokens": 11379813.0,
"step": 6555
},
{
"entropy": 5.628446245193482,
"epoch": 0.5104065356934449,
"grad_norm": 1.2421875,
"learning_rate": 0.0004978932640548794,
"loss": 5.5171,
"mean_token_accuracy": 0.15718891769647597,
"num_tokens": 11388029.0,
"step": 6560
},
{
"entropy": 5.600995826721191,
"epoch": 0.5107955650651624,
"grad_norm": 1.265625,
"learning_rate": 0.0004978894785140429,
"loss": 5.458,
"mean_token_accuracy": 0.15973847061395646,
"num_tokens": 11395852.0,
"step": 6565
},
{
"entropy": 5.605973625183106,
"epoch": 0.51118459443688,
"grad_norm": 1.21875,
"learning_rate": 0.0004978856895912132,
"loss": 5.5467,
"mean_token_accuracy": 0.1593152642250061,
"num_tokens": 11404756.0,
"step": 6570
},
{
"entropy": 5.763776159286499,
"epoch": 0.5115736238085975,
"grad_norm": 1.453125,
"learning_rate": 0.0004978818972864481,
"loss": 5.7037,
"mean_token_accuracy": 0.149703748524189,
"num_tokens": 11413301.0,
"step": 6575
},
{
"entropy": 5.687377214431763,
"epoch": 0.5119626531803151,
"grad_norm": 1.21875,
"learning_rate": 0.0004978781015998048,
"loss": 5.4517,
"mean_token_accuracy": 0.16738782972097396,
"num_tokens": 11421034.0,
"step": 6580
},
{
"entropy": 5.586408901214599,
"epoch": 0.5123516825520327,
"grad_norm": 1.3671875,
"learning_rate": 0.0004978743025313413,
"loss": 5.4916,
"mean_token_accuracy": 0.15781426280736924,
"num_tokens": 11429066.0,
"step": 6585
},
{
"entropy": 5.5836893081665036,
"epoch": 0.5127407119237503,
"grad_norm": 1.2734375,
"learning_rate": 0.0004978705000811148,
"loss": 5.5184,
"mean_token_accuracy": 0.16265126764774324,
"num_tokens": 11438109.0,
"step": 6590
},
{
"entropy": 5.682431221008301,
"epoch": 0.5131297412954678,
"grad_norm": 1.3125,
"learning_rate": 0.0004978666942491832,
"loss": 5.5468,
"mean_token_accuracy": 0.16171859800815583,
"num_tokens": 11446787.0,
"step": 6595
},
{
"entropy": 5.576941347122192,
"epoch": 0.5135187706671853,
"grad_norm": 1.265625,
"learning_rate": 0.0004978628850356043,
"loss": 5.4978,
"mean_token_accuracy": 0.15480518341064453,
"num_tokens": 11456262.0,
"step": 6600
},
{
"entropy": 5.6838137149810795,
"epoch": 0.5139078000389029,
"grad_norm": 1.3046875,
"learning_rate": 0.0004978590724404358,
"loss": 5.4963,
"mean_token_accuracy": 0.16990127712488173,
"num_tokens": 11464297.0,
"step": 6605
},
{
"entropy": 5.693605375289917,
"epoch": 0.5142968294106205,
"grad_norm": 1.2734375,
"learning_rate": 0.0004978552564637356,
"loss": 5.5981,
"mean_token_accuracy": 0.15441921055316926,
"num_tokens": 11473519.0,
"step": 6610
},
{
"entropy": 5.644426012039185,
"epoch": 0.5146858587823381,
"grad_norm": 1.3203125,
"learning_rate": 0.0004978514371055616,
"loss": 5.5747,
"mean_token_accuracy": 0.15011529326438905,
"num_tokens": 11482249.0,
"step": 6615
},
{
"entropy": 5.589239454269409,
"epoch": 0.5150748881540557,
"grad_norm": 1.3515625,
"learning_rate": 0.0004978476143659718,
"loss": 5.5288,
"mean_token_accuracy": 0.1568063020706177,
"num_tokens": 11490687.0,
"step": 6620
},
{
"entropy": 5.607043504714966,
"epoch": 0.5154639175257731,
"grad_norm": 1.265625,
"learning_rate": 0.0004978437882450241,
"loss": 5.5485,
"mean_token_accuracy": 0.16212338656187059,
"num_tokens": 11500225.0,
"step": 6625
},
{
"entropy": 5.6719512939453125,
"epoch": 0.5158529468974907,
"grad_norm": 1.28125,
"learning_rate": 0.0004978399587427766,
"loss": 5.4822,
"mean_token_accuracy": 0.16089026927947997,
"num_tokens": 11508116.0,
"step": 6630
},
{
"entropy": 5.567082118988037,
"epoch": 0.5162419762692083,
"grad_norm": 1.25,
"learning_rate": 0.0004978361258592874,
"loss": 5.5642,
"mean_token_accuracy": 0.15507243424654008,
"num_tokens": 11516525.0,
"step": 6635
},
{
"entropy": 5.6446949481964115,
"epoch": 0.5166310056409259,
"grad_norm": 1.265625,
"learning_rate": 0.0004978322895946147,
"loss": 5.538,
"mean_token_accuracy": 0.16238362938165665,
"num_tokens": 11525434.0,
"step": 6640
},
{
"entropy": 5.695188665390015,
"epoch": 0.5170200350126435,
"grad_norm": 1.15625,
"learning_rate": 0.0004978284499488167,
"loss": 5.5602,
"mean_token_accuracy": 0.15848546028137206,
"num_tokens": 11534318.0,
"step": 6645
},
{
"entropy": 5.600885820388794,
"epoch": 0.5174090643843611,
"grad_norm": 1.2890625,
"learning_rate": 0.0004978246069219516,
"loss": 5.5403,
"mean_token_accuracy": 0.16719124466180801,
"num_tokens": 11542777.0,
"step": 6650
},
{
"entropy": 5.561204528808593,
"epoch": 0.5177980937560785,
"grad_norm": 1.296875,
"learning_rate": 0.0004978207605140777,
"loss": 5.5148,
"mean_token_accuracy": 0.16363105028867722,
"num_tokens": 11550888.0,
"step": 6655
},
{
"entropy": 5.622227907180786,
"epoch": 0.5181871231277961,
"grad_norm": 1.359375,
"learning_rate": 0.0004978169107252534,
"loss": 5.5435,
"mean_token_accuracy": 0.15494584143161774,
"num_tokens": 11560047.0,
"step": 6660
},
{
"entropy": 5.755712890625,
"epoch": 0.5185761524995137,
"grad_norm": 1.4140625,
"learning_rate": 0.0004978130575555373,
"loss": 5.5661,
"mean_token_accuracy": 0.1614314630627632,
"num_tokens": 11568862.0,
"step": 6665
},
{
"entropy": 5.568341636657715,
"epoch": 0.5189651818712313,
"grad_norm": 1.3515625,
"learning_rate": 0.0004978092010049877,
"loss": 5.5043,
"mean_token_accuracy": 0.163470259308815,
"num_tokens": 11577509.0,
"step": 6670
},
{
"entropy": 5.6859883785247805,
"epoch": 0.5193542112429489,
"grad_norm": 1.3828125,
"learning_rate": 0.000497805341073663,
"loss": 5.6282,
"mean_token_accuracy": 0.15651989430189134,
"num_tokens": 11585902.0,
"step": 6675
},
{
"entropy": 5.674084663391113,
"epoch": 0.5197432406146664,
"grad_norm": 1.3125,
"learning_rate": 0.0004978014777616219,
"loss": 5.6123,
"mean_token_accuracy": 0.16017613857984542,
"num_tokens": 11594895.0,
"step": 6680
},
{
"entropy": 5.586768913269043,
"epoch": 0.520132269986384,
"grad_norm": 1.2890625,
"learning_rate": 0.0004977976110689229,
"loss": 5.4369,
"mean_token_accuracy": 0.162335567176342,
"num_tokens": 11604080.0,
"step": 6685
},
{
"entropy": 5.687888431549072,
"epoch": 0.5205212993581015,
"grad_norm": 1.171875,
"learning_rate": 0.000497793740995625,
"loss": 5.5551,
"mean_token_accuracy": 0.15847532749176024,
"num_tokens": 11613370.0,
"step": 6690
},
{
"entropy": 5.589132499694824,
"epoch": 0.5209103287298191,
"grad_norm": 1.3046875,
"learning_rate": 0.0004977898675417866,
"loss": 5.5413,
"mean_token_accuracy": 0.15930339246988295,
"num_tokens": 11621913.0,
"step": 6695
},
{
"entropy": 5.677670526504516,
"epoch": 0.5212993581015367,
"grad_norm": 1.34375,
"learning_rate": 0.0004977859907074664,
"loss": 5.6869,
"mean_token_accuracy": 0.1528049200773239,
"num_tokens": 11630393.0,
"step": 6700
},
{
"entropy": 5.732822036743164,
"epoch": 0.5216883874732542,
"grad_norm": 1.4453125,
"learning_rate": 0.0004977821104927236,
"loss": 5.6297,
"mean_token_accuracy": 0.1573626160621643,
"num_tokens": 11639038.0,
"step": 6705
},
{
"entropy": 5.6175659656524655,
"epoch": 0.5220774168449718,
"grad_norm": 1.296875,
"learning_rate": 0.0004977782268976167,
"loss": 5.5012,
"mean_token_accuracy": 0.15832498222589492,
"num_tokens": 11647355.0,
"step": 6710
},
{
"entropy": 5.581278133392334,
"epoch": 0.5224664462166894,
"grad_norm": 1.359375,
"learning_rate": 0.0004977743399222048,
"loss": 5.5111,
"mean_token_accuracy": 0.15422431975603104,
"num_tokens": 11655899.0,
"step": 6715
},
{
"entropy": 5.753375005722046,
"epoch": 0.522855475588407,
"grad_norm": 1.296875,
"learning_rate": 0.0004977704495665471,
"loss": 5.6935,
"mean_token_accuracy": 0.14948423206806183,
"num_tokens": 11664602.0,
"step": 6720
},
{
"entropy": 5.686633586883545,
"epoch": 0.5232445049601245,
"grad_norm": 1.3125,
"learning_rate": 0.0004977665558307023,
"loss": 5.5188,
"mean_token_accuracy": 0.1614977553486824,
"num_tokens": 11673011.0,
"step": 6725
},
{
"entropy": 5.728595113754272,
"epoch": 0.523633534331842,
"grad_norm": 1.3984375,
"learning_rate": 0.0004977626587147295,
"loss": 5.6987,
"mean_token_accuracy": 0.14548852294683456,
"num_tokens": 11681989.0,
"step": 6730
},
{
"entropy": 5.722722291946411,
"epoch": 0.5240225637035596,
"grad_norm": 1.3046875,
"learning_rate": 0.000497758758218688,
"loss": 5.6269,
"mean_token_accuracy": 0.15139962285757064,
"num_tokens": 11690502.0,
"step": 6735
},
{
"entropy": 5.672935247421265,
"epoch": 0.5244115930752772,
"grad_norm": 1.390625,
"learning_rate": 0.0004977548543426368,
"loss": 5.6226,
"mean_token_accuracy": 0.1551753543317318,
"num_tokens": 11699594.0,
"step": 6740
},
{
"entropy": 5.587326717376709,
"epoch": 0.5248006224469948,
"grad_norm": 1.3984375,
"learning_rate": 0.0004977509470866353,
"loss": 5.4916,
"mean_token_accuracy": 0.16007392406463622,
"num_tokens": 11708009.0,
"step": 6745
},
{
"entropy": 5.707208347320557,
"epoch": 0.5251896518187124,
"grad_norm": 1.46875,
"learning_rate": 0.0004977470364507427,
"loss": 5.6656,
"mean_token_accuracy": 0.1489517107605934,
"num_tokens": 11716592.0,
"step": 6750
},
{
"entropy": 5.694457864761352,
"epoch": 0.5255786811904298,
"grad_norm": 1.3125,
"learning_rate": 0.0004977431224350184,
"loss": 5.6656,
"mean_token_accuracy": 0.1458994671702385,
"num_tokens": 11725894.0,
"step": 6755
},
{
"entropy": 5.768098592758179,
"epoch": 0.5259677105621474,
"grad_norm": 1.421875,
"learning_rate": 0.0004977392050395217,
"loss": 5.5362,
"mean_token_accuracy": 0.1602727875113487,
"num_tokens": 11734141.0,
"step": 6760
},
{
"entropy": 5.642729377746582,
"epoch": 0.526356739933865,
"grad_norm": 1.4296875,
"learning_rate": 0.0004977352842643121,
"loss": 5.6457,
"mean_token_accuracy": 0.15595187842845917,
"num_tokens": 11742539.0,
"step": 6765
},
{
"entropy": 5.648681259155273,
"epoch": 0.5267457693055826,
"grad_norm": 1.3125,
"learning_rate": 0.000497731360109449,
"loss": 5.5973,
"mean_token_accuracy": 0.15659012347459794,
"num_tokens": 11751649.0,
"step": 6770
},
{
"entropy": 5.729363012313843,
"epoch": 0.5271347986773002,
"grad_norm": 1.4453125,
"learning_rate": 0.0004977274325749922,
"loss": 5.6011,
"mean_token_accuracy": 0.1551893211901188,
"num_tokens": 11760376.0,
"step": 6775
},
{
"entropy": 5.703770875930786,
"epoch": 0.5275238280490177,
"grad_norm": 1.3125,
"learning_rate": 0.0004977235016610011,
"loss": 5.564,
"mean_token_accuracy": 0.15711267292499542,
"num_tokens": 11768644.0,
"step": 6780
},
{
"entropy": 5.734229660034179,
"epoch": 0.5279128574207352,
"grad_norm": 1.4375,
"learning_rate": 0.0004977195673675353,
"loss": 5.6425,
"mean_token_accuracy": 0.15375020503997802,
"num_tokens": 11777282.0,
"step": 6785
},
{
"entropy": 5.625796604156494,
"epoch": 0.5283018867924528,
"grad_norm": 1.390625,
"learning_rate": 0.0004977156296946545,
"loss": 5.5393,
"mean_token_accuracy": 0.15840379446744918,
"num_tokens": 11786108.0,
"step": 6790
},
{
"entropy": 5.676083612442016,
"epoch": 0.5286909161641704,
"grad_norm": 1.3671875,
"learning_rate": 0.0004977116886424187,
"loss": 5.5956,
"mean_token_accuracy": 0.15348794162273408,
"num_tokens": 11795151.0,
"step": 6795
},
{
"entropy": 5.786767816543579,
"epoch": 0.529079945535888,
"grad_norm": 1.3984375,
"learning_rate": 0.0004977077442108873,
"loss": 5.5445,
"mean_token_accuracy": 0.16012217849493027,
"num_tokens": 11803971.0,
"step": 6800
},
{
"entropy": 5.579844236373901,
"epoch": 0.5294689749076055,
"grad_norm": 1.4765625,
"learning_rate": 0.0004977037964001204,
"loss": 5.6101,
"mean_token_accuracy": 0.1626015529036522,
"num_tokens": 11812483.0,
"step": 6805
},
{
"entropy": 5.598951911926269,
"epoch": 0.5298580042793231,
"grad_norm": 1.3359375,
"learning_rate": 0.000497699845210178,
"loss": 5.4882,
"mean_token_accuracy": 0.15942364782094956,
"num_tokens": 11821144.0,
"step": 6810
},
{
"entropy": 5.649045038223266,
"epoch": 0.5302470336510406,
"grad_norm": 1.359375,
"learning_rate": 0.0004976958906411197,
"loss": 5.4502,
"mean_token_accuracy": 0.16786729842424392,
"num_tokens": 11829298.0,
"step": 6815
},
{
"entropy": 5.6356611251831055,
"epoch": 0.5306360630227582,
"grad_norm": 1.34375,
"learning_rate": 0.0004976919326930059,
"loss": 5.5413,
"mean_token_accuracy": 0.1632298156619072,
"num_tokens": 11837558.0,
"step": 6820
},
{
"entropy": 5.538313484191894,
"epoch": 0.5310250923944758,
"grad_norm": 1.3046875,
"learning_rate": 0.0004976879713658964,
"loss": 5.4329,
"mean_token_accuracy": 0.16372002512216569,
"num_tokens": 11845770.0,
"step": 6825
},
{
"entropy": 5.641631031036377,
"epoch": 0.5314141217661933,
"grad_norm": 1.3359375,
"learning_rate": 0.0004976840066598514,
"loss": 5.6338,
"mean_token_accuracy": 0.14802222698926926,
"num_tokens": 11854351.0,
"step": 6830
},
{
"entropy": 5.793056392669678,
"epoch": 0.5318031511379109,
"grad_norm": 1.3515625,
"learning_rate": 0.0004976800385749309,
"loss": 5.6301,
"mean_token_accuracy": 0.16096321791410445,
"num_tokens": 11862610.0,
"step": 6835
},
{
"entropy": 5.6363152980804445,
"epoch": 0.5321921805096285,
"grad_norm": 1.34375,
"learning_rate": 0.0004976760671111954,
"loss": 5.5878,
"mean_token_accuracy": 0.16149788051843644,
"num_tokens": 11871078.0,
"step": 6840
},
{
"entropy": 5.602367925643921,
"epoch": 0.532581209881346,
"grad_norm": 1.2734375,
"learning_rate": 0.000497672092268705,
"loss": 5.4558,
"mean_token_accuracy": 0.16823557317256926,
"num_tokens": 11879303.0,
"step": 6845
},
{
"entropy": 5.697409915924072,
"epoch": 0.5329702392530636,
"grad_norm": 1.484375,
"learning_rate": 0.00049766811404752,
"loss": 5.5198,
"mean_token_accuracy": 0.15866248458623886,
"num_tokens": 11887855.0,
"step": 6850
},
{
"entropy": 5.669441270828247,
"epoch": 0.5333592686247812,
"grad_norm": 1.234375,
"learning_rate": 0.0004976641324477008,
"loss": 5.4896,
"mean_token_accuracy": 0.1634063646197319,
"num_tokens": 11896262.0,
"step": 6855
},
{
"entropy": 5.4838920593261715,
"epoch": 0.5337482979964987,
"grad_norm": 1.375,
"learning_rate": 0.0004976601474693077,
"loss": 5.4028,
"mean_token_accuracy": 0.16620713621377944,
"num_tokens": 11905140.0,
"step": 6860
},
{
"entropy": 5.534974670410156,
"epoch": 0.5341373273682163,
"grad_norm": 1.4375,
"learning_rate": 0.0004976561591124014,
"loss": 5.4434,
"mean_token_accuracy": 0.1693867936730385,
"num_tokens": 11913177.0,
"step": 6865
},
{
"entropy": 5.501511096954346,
"epoch": 0.5345263567399339,
"grad_norm": 1.359375,
"learning_rate": 0.0004976521673770421,
"loss": 5.4746,
"mean_token_accuracy": 0.16013596653938295,
"num_tokens": 11921955.0,
"step": 6870
},
{
"entropy": 5.614723873138428,
"epoch": 0.5349153861116515,
"grad_norm": 1.328125,
"learning_rate": 0.0004976481722632907,
"loss": 5.484,
"mean_token_accuracy": 0.15625718086957932,
"num_tokens": 11930976.0,
"step": 6875
},
{
"entropy": 5.621336030960083,
"epoch": 0.535304415483369,
"grad_norm": 1.2421875,
"learning_rate": 0.0004976441737712076,
"loss": 5.4153,
"mean_token_accuracy": 0.16089983880519867,
"num_tokens": 11939708.0,
"step": 6880
},
{
"entropy": 5.668586921691895,
"epoch": 0.5356934448550865,
"grad_norm": 1.3984375,
"learning_rate": 0.0004976401719008536,
"loss": 5.6059,
"mean_token_accuracy": 0.15665263086557388,
"num_tokens": 11949176.0,
"step": 6885
},
{
"entropy": 5.609954023361206,
"epoch": 0.5360824742268041,
"grad_norm": 1.2421875,
"learning_rate": 0.0004976361666522893,
"loss": 5.4534,
"mean_token_accuracy": 0.16610209494829178,
"num_tokens": 11958089.0,
"step": 6890
},
{
"entropy": 5.6360095024108885,
"epoch": 0.5364715035985217,
"grad_norm": 1.75,
"learning_rate": 0.0004976321580255755,
"loss": 5.5369,
"mean_token_accuracy": 0.16299730837345122,
"num_tokens": 11966719.0,
"step": 6895
},
{
"entropy": 5.68487663269043,
"epoch": 0.5368605329702393,
"grad_norm": 1.28125,
"learning_rate": 0.0004976281460207731,
"loss": 5.6002,
"mean_token_accuracy": 0.15535397827625275,
"num_tokens": 11975959.0,
"step": 6900
},
{
"entropy": 5.674017238616943,
"epoch": 0.5372495623419569,
"grad_norm": 1.5234375,
"learning_rate": 0.000497624130637943,
"loss": 5.5641,
"mean_token_accuracy": 0.16273371428251265,
"num_tokens": 11984282.0,
"step": 6905
},
{
"entropy": 5.644144487380982,
"epoch": 0.5376385917136743,
"grad_norm": 1.34375,
"learning_rate": 0.000497620111877146,
"loss": 5.5637,
"mean_token_accuracy": 0.15435383319854737,
"num_tokens": 11993289.0,
"step": 6910
},
{
"entropy": 5.60618839263916,
"epoch": 0.5380276210853919,
"grad_norm": 1.375,
"learning_rate": 0.0004976160897384431,
"loss": 5.4969,
"mean_token_accuracy": 0.16079047173261643,
"num_tokens": 12002425.0,
"step": 6915
},
{
"entropy": 5.638632774353027,
"epoch": 0.5384166504571095,
"grad_norm": 1.296875,
"learning_rate": 0.0004976120642218955,
"loss": 5.5134,
"mean_token_accuracy": 0.15918543487787246,
"num_tokens": 12010435.0,
"step": 6920
},
{
"entropy": 5.5793486595153805,
"epoch": 0.5388056798288271,
"grad_norm": 1.4296875,
"learning_rate": 0.000497608035327564,
"loss": 5.5623,
"mean_token_accuracy": 0.15495380908250808,
"num_tokens": 12018768.0,
"step": 6925
},
{
"entropy": 5.678703308105469,
"epoch": 0.5391947092005447,
"grad_norm": 1.3046875,
"learning_rate": 0.00049760400305551,
"loss": 5.5672,
"mean_token_accuracy": 0.1580612525343895,
"num_tokens": 12027617.0,
"step": 6930
},
{
"entropy": 5.695169067382812,
"epoch": 0.5395837385722622,
"grad_norm": 1.359375,
"learning_rate": 0.0004975999674057944,
"loss": 5.5489,
"mean_token_accuracy": 0.15605789721012114,
"num_tokens": 12036024.0,
"step": 6935
},
{
"entropy": 5.723021030426025,
"epoch": 0.5399727679439797,
"grad_norm": 1.296875,
"learning_rate": 0.0004975959283784787,
"loss": 5.5789,
"mean_token_accuracy": 0.1569458082318306,
"num_tokens": 12044206.0,
"step": 6940
},
{
"entropy": 5.619761753082275,
"epoch": 0.5403617973156973,
"grad_norm": 1.1796875,
"learning_rate": 0.0004975918859736241,
"loss": 5.4611,
"mean_token_accuracy": 0.16328767985105513,
"num_tokens": 12052294.0,
"step": 6945
},
{
"entropy": 5.593639135360718,
"epoch": 0.5407508266874149,
"grad_norm": 1.3203125,
"learning_rate": 0.0004975878401912919,
"loss": 5.5733,
"mean_token_accuracy": 0.15376802980899812,
"num_tokens": 12060446.0,
"step": 6950
},
{
"entropy": 5.663234519958496,
"epoch": 0.5411398560591325,
"grad_norm": 1.3359375,
"learning_rate": 0.0004975837910315436,
"loss": 5.6616,
"mean_token_accuracy": 0.15635811388492585,
"num_tokens": 12069267.0,
"step": 6955
},
{
"entropy": 5.709695672988891,
"epoch": 0.54152888543085,
"grad_norm": 1.2734375,
"learning_rate": 0.0004975797384944405,
"loss": 5.5824,
"mean_token_accuracy": 0.15533999651670455,
"num_tokens": 12077808.0,
"step": 6960
},
{
"entropy": 5.638394355773926,
"epoch": 0.5419179148025676,
"grad_norm": 1.3359375,
"learning_rate": 0.000497575682580044,
"loss": 5.5542,
"mean_token_accuracy": 0.1539350613951683,
"num_tokens": 12086893.0,
"step": 6965
},
{
"entropy": 5.666961669921875,
"epoch": 0.5423069441742852,
"grad_norm": 1.28125,
"learning_rate": 0.000497571623288416,
"loss": 5.4399,
"mean_token_accuracy": 0.1644754961133003,
"num_tokens": 12095462.0,
"step": 6970
},
{
"entropy": 5.6487678527832035,
"epoch": 0.5426959735460027,
"grad_norm": 1.4140625,
"learning_rate": 0.0004975675606196177,
"loss": 5.5295,
"mean_token_accuracy": 0.15933847576379775,
"num_tokens": 12103477.0,
"step": 6975
},
{
"entropy": 5.671709299087524,
"epoch": 0.5430850029177203,
"grad_norm": 1.46875,
"learning_rate": 0.000497563494573711,
"loss": 5.6699,
"mean_token_accuracy": 0.15767599493265153,
"num_tokens": 12112752.0,
"step": 6980
},
{
"entropy": 5.7197895526885985,
"epoch": 0.5434740322894378,
"grad_norm": 1.3984375,
"learning_rate": 0.0004975594251507575,
"loss": 5.5767,
"mean_token_accuracy": 0.1488715760409832,
"num_tokens": 12121883.0,
"step": 6985
},
{
"entropy": 5.684212684631348,
"epoch": 0.5438630616611554,
"grad_norm": 1.3515625,
"learning_rate": 0.000497555352350819,
"loss": 5.5251,
"mean_token_accuracy": 0.15814283341169358,
"num_tokens": 12130316.0,
"step": 6990
},
{
"entropy": 5.640691900253296,
"epoch": 0.544252091032873,
"grad_norm": 1.265625,
"learning_rate": 0.0004975512761739572,
"loss": 5.5032,
"mean_token_accuracy": 0.15524298548698426,
"num_tokens": 12138856.0,
"step": 6995
},
{
"entropy": 5.592985916137695,
"epoch": 0.5446411204045906,
"grad_norm": 1.390625,
"learning_rate": 0.000497547196620234,
"loss": 5.5785,
"mean_token_accuracy": 0.15525238662958146,
"num_tokens": 12147933.0,
"step": 7000
},
{
"entropy": 5.633612012863159,
"epoch": 0.5450301497763081,
"grad_norm": 1.3046875,
"learning_rate": 0.0004975431136897114,
"loss": 5.4554,
"mean_token_accuracy": 0.1621299922466278,
"num_tokens": 12156529.0,
"step": 7005
},
{
"entropy": 5.597944164276123,
"epoch": 0.5454191791480256,
"grad_norm": 1.296875,
"learning_rate": 0.0004975390273824512,
"loss": 5.4113,
"mean_token_accuracy": 0.16376460194587708,
"num_tokens": 12165233.0,
"step": 7010
},
{
"entropy": 5.611815452575684,
"epoch": 0.5458082085197432,
"grad_norm": 1.3203125,
"learning_rate": 0.0004975349376985155,
"loss": 5.6011,
"mean_token_accuracy": 0.15549773424863816,
"num_tokens": 12173260.0,
"step": 7015
},
{
"entropy": 5.673900985717774,
"epoch": 0.5461972378914608,
"grad_norm": 1.328125,
"learning_rate": 0.0004975308446379663,
"loss": 5.6101,
"mean_token_accuracy": 0.15461681336164473,
"num_tokens": 12181676.0,
"step": 7020
},
{
"entropy": 5.6847456932067875,
"epoch": 0.5465862672631784,
"grad_norm": 1.40625,
"learning_rate": 0.0004975267482008657,
"loss": 5.5254,
"mean_token_accuracy": 0.1594758614897728,
"num_tokens": 12189485.0,
"step": 7025
},
{
"entropy": 5.6035644054412845,
"epoch": 0.546975296634896,
"grad_norm": 1.3046875,
"learning_rate": 0.000497522648387276,
"loss": 5.4527,
"mean_token_accuracy": 0.16365523785352706,
"num_tokens": 12198429.0,
"step": 7030
},
{
"entropy": 5.577117395401001,
"epoch": 0.5473643260066134,
"grad_norm": 1.40625,
"learning_rate": 0.0004975185451972592,
"loss": 5.572,
"mean_token_accuracy": 0.16115963011980056,
"num_tokens": 12207404.0,
"step": 7035
},
{
"entropy": 5.701930236816406,
"epoch": 0.547753355378331,
"grad_norm": 1.2421875,
"learning_rate": 0.0004975144386308776,
"loss": 5.5479,
"mean_token_accuracy": 0.15822896361351013,
"num_tokens": 12216301.0,
"step": 7040
},
{
"entropy": 5.591483783721924,
"epoch": 0.5481423847500486,
"grad_norm": 1.3359375,
"learning_rate": 0.0004975103286881936,
"loss": 5.5384,
"mean_token_accuracy": 0.16383027136325837,
"num_tokens": 12224873.0,
"step": 7045
},
{
"entropy": 5.612608337402344,
"epoch": 0.5485314141217662,
"grad_norm": 1.2890625,
"learning_rate": 0.0004975062153692696,
"loss": 5.5158,
"mean_token_accuracy": 0.1623031973838806,
"num_tokens": 12233439.0,
"step": 7050
},
{
"entropy": 5.694883489608765,
"epoch": 0.5489204434934838,
"grad_norm": 1.390625,
"learning_rate": 0.0004975020986741678,
"loss": 5.5703,
"mean_token_accuracy": 0.15239009335637094,
"num_tokens": 12242235.0,
"step": 7055
},
{
"entropy": 5.609505271911621,
"epoch": 0.5493094728652014,
"grad_norm": 1.4296875,
"learning_rate": 0.0004974979786029509,
"loss": 5.5242,
"mean_token_accuracy": 0.16340847909450532,
"num_tokens": 12251295.0,
"step": 7060
},
{
"entropy": 5.638507509231568,
"epoch": 0.5496985022369189,
"grad_norm": 1.34375,
"learning_rate": 0.0004974938551556814,
"loss": 5.4621,
"mean_token_accuracy": 0.1659865766763687,
"num_tokens": 12260130.0,
"step": 7065
},
{
"entropy": 5.6084636688232425,
"epoch": 0.5500875316086364,
"grad_norm": 1.265625,
"learning_rate": 0.0004974897283324217,
"loss": 5.5148,
"mean_token_accuracy": 0.160197813808918,
"num_tokens": 12269542.0,
"step": 7070
},
{
"entropy": 5.6162344455719,
"epoch": 0.550476560980354,
"grad_norm": 1.3515625,
"learning_rate": 0.0004974855981332343,
"loss": 5.6004,
"mean_token_accuracy": 0.16120688617229462,
"num_tokens": 12278174.0,
"step": 7075
},
{
"entropy": 5.673370409011841,
"epoch": 0.5508655903520716,
"grad_norm": 1.3203125,
"learning_rate": 0.0004974814645581823,
"loss": 5.6678,
"mean_token_accuracy": 0.15251821428537368,
"num_tokens": 12287651.0,
"step": 7080
},
{
"entropy": 5.763124084472656,
"epoch": 0.5512546197237892,
"grad_norm": 1.3515625,
"learning_rate": 0.0004974773276073282,
"loss": 5.6438,
"mean_token_accuracy": 0.15593429803848266,
"num_tokens": 12296009.0,
"step": 7085
},
{
"entropy": 5.612685346603394,
"epoch": 0.5516436490955067,
"grad_norm": 1.390625,
"learning_rate": 0.0004974731872807347,
"loss": 5.5443,
"mean_token_accuracy": 0.15328624844551086,
"num_tokens": 12305406.0,
"step": 7090
},
{
"entropy": 5.6309099197387695,
"epoch": 0.5520326784672243,
"grad_norm": 1.296875,
"learning_rate": 0.0004974690435784647,
"loss": 5.5797,
"mean_token_accuracy": 0.15415759533643722,
"num_tokens": 12314169.0,
"step": 7095
},
{
"entropy": 5.598780250549316,
"epoch": 0.5524217078389418,
"grad_norm": 1.3984375,
"learning_rate": 0.0004974648965005811,
"loss": 5.4733,
"mean_token_accuracy": 0.16661957502365113,
"num_tokens": 12323019.0,
"step": 7100
},
{
"entropy": 5.621094560623169,
"epoch": 0.5528107372106594,
"grad_norm": 1.5390625,
"learning_rate": 0.0004974607460471466,
"loss": 5.5827,
"mean_token_accuracy": 0.1506951481103897,
"num_tokens": 12331491.0,
"step": 7105
},
{
"entropy": 5.62510232925415,
"epoch": 0.553199766582377,
"grad_norm": 1.40625,
"learning_rate": 0.0004974565922182246,
"loss": 5.5438,
"mean_token_accuracy": 0.15989115089178085,
"num_tokens": 12339647.0,
"step": 7110
},
{
"entropy": 5.607163286209106,
"epoch": 0.5535887959540945,
"grad_norm": 1.484375,
"learning_rate": 0.0004974524350138777,
"loss": 5.522,
"mean_token_accuracy": 0.15979181230068207,
"num_tokens": 12348469.0,
"step": 7115
},
{
"entropy": 5.6572285175323485,
"epoch": 0.5539778253258121,
"grad_norm": 1.40625,
"learning_rate": 0.0004974482744341693,
"loss": 5.4974,
"mean_token_accuracy": 0.16333761662244797,
"num_tokens": 12357168.0,
"step": 7120
},
{
"entropy": 5.534380054473877,
"epoch": 0.5543668546975297,
"grad_norm": 1.3828125,
"learning_rate": 0.0004974441104791624,
"loss": 5.4967,
"mean_token_accuracy": 0.1589392438530922,
"num_tokens": 12365750.0,
"step": 7125
},
{
"entropy": 5.601346969604492,
"epoch": 0.5547558840692473,
"grad_norm": 1.515625,
"learning_rate": 0.0004974399431489201,
"loss": 5.5427,
"mean_token_accuracy": 0.15742712914943696,
"num_tokens": 12374086.0,
"step": 7130
},
{
"entropy": 5.660343980789184,
"epoch": 0.5551449134409648,
"grad_norm": 1.3828125,
"learning_rate": 0.0004974357724435057,
"loss": 5.4835,
"mean_token_accuracy": 0.16328005641698837,
"num_tokens": 12381997.0,
"step": 7135
},
{
"entropy": 5.676285028457642,
"epoch": 0.5555339428126823,
"grad_norm": 1.3515625,
"learning_rate": 0.0004974315983629825,
"loss": 5.5817,
"mean_token_accuracy": 0.1614387571811676,
"num_tokens": 12391083.0,
"step": 7140
},
{
"entropy": 5.644300699234009,
"epoch": 0.5559229721843999,
"grad_norm": 1.3984375,
"learning_rate": 0.0004974274209074139,
"loss": 5.497,
"mean_token_accuracy": 0.16053244918584825,
"num_tokens": 12400016.0,
"step": 7145
},
{
"entropy": 5.698569917678833,
"epoch": 0.5563120015561175,
"grad_norm": 1.3828125,
"learning_rate": 0.0004974232400768632,
"loss": 5.6068,
"mean_token_accuracy": 0.15869893729686738,
"num_tokens": 12409255.0,
"step": 7150
},
{
"entropy": 5.660106754302978,
"epoch": 0.5567010309278351,
"grad_norm": 1.2890625,
"learning_rate": 0.0004974190558713939,
"loss": 5.4848,
"mean_token_accuracy": 0.16400493532419205,
"num_tokens": 12417765.0,
"step": 7155
},
{
"entropy": 5.702695846557617,
"epoch": 0.5570900602995527,
"grad_norm": 1.3203125,
"learning_rate": 0.0004974148682910694,
"loss": 5.7475,
"mean_token_accuracy": 0.1424033060669899,
"num_tokens": 12427588.0,
"step": 7160
},
{
"entropy": 5.648605871200561,
"epoch": 0.5574790896712701,
"grad_norm": 1.3671875,
"learning_rate": 0.0004974106773359533,
"loss": 5.4138,
"mean_token_accuracy": 0.16430372297763823,
"num_tokens": 12435773.0,
"step": 7165
},
{
"entropy": 5.619985103607178,
"epoch": 0.5578681190429877,
"grad_norm": 1.7421875,
"learning_rate": 0.0004974064830061091,
"loss": 5.5464,
"mean_token_accuracy": 0.16402772665023804,
"num_tokens": 12443837.0,
"step": 7170
},
{
"entropy": 5.6071559429168705,
"epoch": 0.5582571484147053,
"grad_norm": 1.375,
"learning_rate": 0.0004974022853016006,
"loss": 5.5454,
"mean_token_accuracy": 0.16198388189077378,
"num_tokens": 12452945.0,
"step": 7175
},
{
"entropy": 5.632401609420777,
"epoch": 0.5586461777864229,
"grad_norm": 1.6171875,
"learning_rate": 0.0004973980842224913,
"loss": 5.475,
"mean_token_accuracy": 0.16739387661218644,
"num_tokens": 12461008.0,
"step": 7180
},
{
"entropy": 5.5655022144317625,
"epoch": 0.5590352071581405,
"grad_norm": 1.3984375,
"learning_rate": 0.0004973938797688452,
"loss": 5.4734,
"mean_token_accuracy": 0.156940957903862,
"num_tokens": 12469314.0,
"step": 7185
},
{
"entropy": 5.636973476409912,
"epoch": 0.559424236529858,
"grad_norm": 1.5625,
"learning_rate": 0.0004973896719407259,
"loss": 5.4887,
"mean_token_accuracy": 0.1571099802851677,
"num_tokens": 12477540.0,
"step": 7190
},
{
"entropy": 5.62240891456604,
"epoch": 0.5598132659015755,
"grad_norm": 1.4296875,
"learning_rate": 0.0004973854607381972,
"loss": 5.5421,
"mean_token_accuracy": 0.16325449496507644,
"num_tokens": 12486894.0,
"step": 7195
},
{
"entropy": 5.655526065826416,
"epoch": 0.5602022952732931,
"grad_norm": 1.375,
"learning_rate": 0.0004973812461613232,
"loss": 5.5413,
"mean_token_accuracy": 0.15660593509674073,
"num_tokens": 12496213.0,
"step": 7200
},
{
"entropy": 5.592194223403931,
"epoch": 0.5605913246450107,
"grad_norm": 1.328125,
"learning_rate": 0.0004973770282101677,
"loss": 5.5315,
"mean_token_accuracy": 0.15898202806711198,
"num_tokens": 12504160.0,
"step": 7205
},
{
"entropy": 5.54472541809082,
"epoch": 0.5609803540167283,
"grad_norm": 1.4921875,
"learning_rate": 0.0004973728068847947,
"loss": 5.474,
"mean_token_accuracy": 0.16442665830254555,
"num_tokens": 12511915.0,
"step": 7210
},
{
"entropy": 5.639721298217774,
"epoch": 0.5613693833884458,
"grad_norm": 1.328125,
"learning_rate": 0.0004973685821852684,
"loss": 5.5125,
"mean_token_accuracy": 0.159575916826725,
"num_tokens": 12520400.0,
"step": 7215
},
{
"entropy": 5.580363988876343,
"epoch": 0.5617584127601634,
"grad_norm": 1.2734375,
"learning_rate": 0.0004973643541116528,
"loss": 5.4226,
"mean_token_accuracy": 0.168043652176857,
"num_tokens": 12528724.0,
"step": 7220
},
{
"entropy": 5.648554515838623,
"epoch": 0.562147442131881,
"grad_norm": 1.3203125,
"learning_rate": 0.000497360122664012,
"loss": 5.6009,
"mean_token_accuracy": 0.15417485386133195,
"num_tokens": 12537378.0,
"step": 7225
},
{
"entropy": 5.626233673095703,
"epoch": 0.5625364715035985,
"grad_norm": 1.234375,
"learning_rate": 0.0004973558878424104,
"loss": 5.5328,
"mean_token_accuracy": 0.16310097873210908,
"num_tokens": 12546080.0,
"step": 7230
},
{
"entropy": 5.6941893100738525,
"epoch": 0.5629255008753161,
"grad_norm": 1.2734375,
"learning_rate": 0.0004973516496469119,
"loss": 5.5547,
"mean_token_accuracy": 0.16019777953624725,
"num_tokens": 12554526.0,
"step": 7235
},
{
"entropy": 5.63044319152832,
"epoch": 0.5633145302470336,
"grad_norm": 1.5390625,
"learning_rate": 0.0004973474080775811,
"loss": 5.5093,
"mean_token_accuracy": 0.16128407865762712,
"num_tokens": 12563018.0,
"step": 7240
},
{
"entropy": 5.627581405639648,
"epoch": 0.5637035596187512,
"grad_norm": 1.3984375,
"learning_rate": 0.0004973431631344824,
"loss": 5.567,
"mean_token_accuracy": 0.1591254025697708,
"num_tokens": 12571611.0,
"step": 7245
},
{
"entropy": 5.698124742507934,
"epoch": 0.5640925889904688,
"grad_norm": 1.2421875,
"learning_rate": 0.00049733891481768,
"loss": 5.635,
"mean_token_accuracy": 0.15900451242923735,
"num_tokens": 12580168.0,
"step": 7250
},
{
"entropy": 5.620501375198364,
"epoch": 0.5644816183621864,
"grad_norm": 1.3203125,
"learning_rate": 0.0004973346631272384,
"loss": 5.5396,
"mean_token_accuracy": 0.1586569592356682,
"num_tokens": 12589843.0,
"step": 7255
},
{
"entropy": 5.712000942230224,
"epoch": 0.5648706477339039,
"grad_norm": 1.3515625,
"learning_rate": 0.0004973304080632222,
"loss": 5.5776,
"mean_token_accuracy": 0.1545659400522709,
"num_tokens": 12598219.0,
"step": 7260
},
{
"entropy": 5.625898599624634,
"epoch": 0.5652596771056215,
"grad_norm": 1.359375,
"learning_rate": 0.000497326149625696,
"loss": 5.5644,
"mean_token_accuracy": 0.16191482096910476,
"num_tokens": 12606665.0,
"step": 7265
},
{
"entropy": 5.556368732452393,
"epoch": 0.565648706477339,
"grad_norm": 1.5625,
"learning_rate": 0.0004973218878147244,
"loss": 5.4893,
"mean_token_accuracy": 0.16165162175893782,
"num_tokens": 12615062.0,
"step": 7270
},
{
"entropy": 5.701360082626342,
"epoch": 0.5660377358490566,
"grad_norm": 1.3125,
"learning_rate": 0.0004973176226303719,
"loss": 5.5805,
"mean_token_accuracy": 0.1605806604027748,
"num_tokens": 12623923.0,
"step": 7275
},
{
"entropy": 5.665620994567871,
"epoch": 0.5664267652207742,
"grad_norm": 1.4765625,
"learning_rate": 0.0004973133540727033,
"loss": 5.4939,
"mean_token_accuracy": 0.15362900421023368,
"num_tokens": 12632579.0,
"step": 7280
},
{
"entropy": 5.6012444496154785,
"epoch": 0.5668157945924918,
"grad_norm": 1.484375,
"learning_rate": 0.0004973090821417835,
"loss": 5.4818,
"mean_token_accuracy": 0.161869415640831,
"num_tokens": 12640847.0,
"step": 7285
},
{
"entropy": 5.6269035816192625,
"epoch": 0.5672048239642093,
"grad_norm": 1.3828125,
"learning_rate": 0.0004973048068376772,
"loss": 5.4605,
"mean_token_accuracy": 0.1663483962416649,
"num_tokens": 12649020.0,
"step": 7290
},
{
"entropy": 5.498252439498901,
"epoch": 0.5675938533359268,
"grad_norm": 1.25,
"learning_rate": 0.0004973005281604492,
"loss": 5.3791,
"mean_token_accuracy": 0.16371892839670182,
"num_tokens": 12657569.0,
"step": 7295
},
{
"entropy": 5.70664758682251,
"epoch": 0.5679828827076444,
"grad_norm": 1.3984375,
"learning_rate": 0.0004972962461101646,
"loss": 5.6503,
"mean_token_accuracy": 0.14907950535416603,
"num_tokens": 12666681.0,
"step": 7300
},
{
"entropy": 5.640994834899902,
"epoch": 0.568371912079362,
"grad_norm": 1.2890625,
"learning_rate": 0.0004972919606868883,
"loss": 5.4688,
"mean_token_accuracy": 0.1601715177297592,
"num_tokens": 12675210.0,
"step": 7305
},
{
"entropy": 5.625789308547974,
"epoch": 0.5687609414510796,
"grad_norm": 1.3515625,
"learning_rate": 0.0004972876718906852,
"loss": 5.5133,
"mean_token_accuracy": 0.1565352126955986,
"num_tokens": 12684019.0,
"step": 7310
},
{
"entropy": 5.622615385055542,
"epoch": 0.5691499708227972,
"grad_norm": 1.359375,
"learning_rate": 0.0004972833797216206,
"loss": 5.5633,
"mean_token_accuracy": 0.14711394608020784,
"num_tokens": 12692891.0,
"step": 7315
},
{
"entropy": 5.657730054855347,
"epoch": 0.5695390001945146,
"grad_norm": 1.640625,
"learning_rate": 0.0004972790841797595,
"loss": 5.4749,
"mean_token_accuracy": 0.16254037916660308,
"num_tokens": 12700606.0,
"step": 7320
},
{
"entropy": 5.584020566940308,
"epoch": 0.5699280295662322,
"grad_norm": 1.546875,
"learning_rate": 0.000497274785265167,
"loss": 5.5761,
"mean_token_accuracy": 0.15465729534626008,
"num_tokens": 12709116.0,
"step": 7325
},
{
"entropy": 5.620603036880493,
"epoch": 0.5703170589379498,
"grad_norm": 1.578125,
"learning_rate": 0.0004972704829779086,
"loss": 5.515,
"mean_token_accuracy": 0.16086004823446273,
"num_tokens": 12717484.0,
"step": 7330
},
{
"entropy": 5.5484014511108395,
"epoch": 0.5707060883096674,
"grad_norm": 1.3828125,
"learning_rate": 0.0004972661773180492,
"loss": 5.4495,
"mean_token_accuracy": 0.1614287480711937,
"num_tokens": 12726216.0,
"step": 7335
},
{
"entropy": 5.596465110778809,
"epoch": 0.571095117681385,
"grad_norm": 1.65625,
"learning_rate": 0.0004972618682856545,
"loss": 5.446,
"mean_token_accuracy": 0.1665715456008911,
"num_tokens": 12735312.0,
"step": 7340
},
{
"entropy": 5.606229066848755,
"epoch": 0.5714841470531025,
"grad_norm": 1.234375,
"learning_rate": 0.0004972575558807896,
"loss": 5.5775,
"mean_token_accuracy": 0.15633748471736908,
"num_tokens": 12744641.0,
"step": 7345
},
{
"entropy": 5.665216493606567,
"epoch": 0.57187317642482,
"grad_norm": 1.3125,
"learning_rate": 0.0004972532401035201,
"loss": 5.5346,
"mean_token_accuracy": 0.15907908231019974,
"num_tokens": 12753951.0,
"step": 7350
},
{
"entropy": 5.723617362976074,
"epoch": 0.5722622057965376,
"grad_norm": 1.4296875,
"learning_rate": 0.0004972489209539114,
"loss": 5.5512,
"mean_token_accuracy": 0.16019677892327308,
"num_tokens": 12762571.0,
"step": 7355
},
{
"entropy": 5.712904691696167,
"epoch": 0.5726512351682552,
"grad_norm": 1.2109375,
"learning_rate": 0.0004972445984320292,
"loss": 5.6205,
"mean_token_accuracy": 0.15869905799627304,
"num_tokens": 12771275.0,
"step": 7360
},
{
"entropy": 5.674981260299683,
"epoch": 0.5730402645399728,
"grad_norm": 1.3203125,
"learning_rate": 0.0004972402725379388,
"loss": 5.5153,
"mean_token_accuracy": 0.1596535786986351,
"num_tokens": 12780488.0,
"step": 7365
},
{
"entropy": 5.6578387260437015,
"epoch": 0.5734292939116903,
"grad_norm": 1.34375,
"learning_rate": 0.000497235943271706,
"loss": 5.5456,
"mean_token_accuracy": 0.1624777302145958,
"num_tokens": 12788588.0,
"step": 7370
},
{
"entropy": 5.560195541381836,
"epoch": 0.5738183232834079,
"grad_norm": 1.4140625,
"learning_rate": 0.0004972316106333966,
"loss": 5.4871,
"mean_token_accuracy": 0.1583039477467537,
"num_tokens": 12797010.0,
"step": 7375
},
{
"entropy": 5.672476291656494,
"epoch": 0.5742073526551255,
"grad_norm": 1.2578125,
"learning_rate": 0.0004972272746230761,
"loss": 5.5737,
"mean_token_accuracy": 0.16159822195768356,
"num_tokens": 12805430.0,
"step": 7380
},
{
"entropy": 5.699739456176758,
"epoch": 0.574596382026843,
"grad_norm": 1.3984375,
"learning_rate": 0.0004972229352408104,
"loss": 5.5415,
"mean_token_accuracy": 0.16303423792123795,
"num_tokens": 12813912.0,
"step": 7385
},
{
"entropy": 5.688510894775391,
"epoch": 0.5749854113985606,
"grad_norm": 1.2890625,
"learning_rate": 0.0004972185924866655,
"loss": 5.5546,
"mean_token_accuracy": 0.15873103514313697,
"num_tokens": 12822421.0,
"step": 7390
},
{
"entropy": 5.554236936569214,
"epoch": 0.5753744407702781,
"grad_norm": 1.3671875,
"learning_rate": 0.0004972142463607071,
"loss": 5.4408,
"mean_token_accuracy": 0.16305190920829774,
"num_tokens": 12830670.0,
"step": 7395
},
{
"entropy": 5.625152826309204,
"epoch": 0.5757634701419957,
"grad_norm": 1.3046875,
"learning_rate": 0.0004972098968630012,
"loss": 5.5553,
"mean_token_accuracy": 0.16401876807212828,
"num_tokens": 12839887.0,
"step": 7400
},
{
"entropy": 5.5732255458831785,
"epoch": 0.5761524995137133,
"grad_norm": 1.2578125,
"learning_rate": 0.0004972055439936137,
"loss": 5.4636,
"mean_token_accuracy": 0.16273102313280105,
"num_tokens": 12848028.0,
"step": 7405
},
{
"entropy": 5.571136522293091,
"epoch": 0.5765415288854309,
"grad_norm": 1.2109375,
"learning_rate": 0.0004972011877526109,
"loss": 5.4718,
"mean_token_accuracy": 0.1594846174120903,
"num_tokens": 12857248.0,
"step": 7410
},
{
"entropy": 5.621489238739014,
"epoch": 0.5769305582571485,
"grad_norm": 1.2890625,
"learning_rate": 0.0004971968281400587,
"loss": 5.4828,
"mean_token_accuracy": 0.1662975862622261,
"num_tokens": 12866352.0,
"step": 7415
},
{
"entropy": 5.666937637329101,
"epoch": 0.5773195876288659,
"grad_norm": 1.671875,
"learning_rate": 0.0004971924651560233,
"loss": 5.5766,
"mean_token_accuracy": 0.15306324139237404,
"num_tokens": 12874548.0,
"step": 7420
},
{
"entropy": 5.623528480529785,
"epoch": 0.5777086170005835,
"grad_norm": 1.609375,
"learning_rate": 0.0004971880988005709,
"loss": 5.3968,
"mean_token_accuracy": 0.16885466426610946,
"num_tokens": 12883172.0,
"step": 7425
},
{
"entropy": 5.517974090576172,
"epoch": 0.5780976463723011,
"grad_norm": 1.40625,
"learning_rate": 0.0004971837290737677,
"loss": 5.48,
"mean_token_accuracy": 0.16351478695869445,
"num_tokens": 12891776.0,
"step": 7430
},
{
"entropy": 5.5739844799041744,
"epoch": 0.5784866757440187,
"grad_norm": 1.3515625,
"learning_rate": 0.0004971793559756801,
"loss": 5.4198,
"mean_token_accuracy": 0.16615208461880684,
"num_tokens": 12900530.0,
"step": 7435
},
{
"entropy": 5.5704282283782955,
"epoch": 0.5788757051157363,
"grad_norm": 1.3671875,
"learning_rate": 0.0004971749795063744,
"loss": 5.4675,
"mean_token_accuracy": 0.16340652704238892,
"num_tokens": 12909260.0,
"step": 7440
},
{
"entropy": 5.615681982040405,
"epoch": 0.5792647344874537,
"grad_norm": 1.328125,
"learning_rate": 0.000497170599665917,
"loss": 5.4631,
"mean_token_accuracy": 0.15883395224809646,
"num_tokens": 12918143.0,
"step": 7445
},
{
"entropy": 5.6724207401275635,
"epoch": 0.5796537638591713,
"grad_norm": 1.421875,
"learning_rate": 0.0004971662164543743,
"loss": 5.5346,
"mean_token_accuracy": 0.16386547535657883,
"num_tokens": 12926361.0,
"step": 7450
},
{
"entropy": 5.593789339065552,
"epoch": 0.5800427932308889,
"grad_norm": 1.53125,
"learning_rate": 0.0004971618298718131,
"loss": 5.4645,
"mean_token_accuracy": 0.16232849061489105,
"num_tokens": 12935080.0,
"step": 7455
},
{
"entropy": 5.644500350952148,
"epoch": 0.5804318226026065,
"grad_norm": 1.4453125,
"learning_rate": 0.0004971574399182995,
"loss": 5.6323,
"mean_token_accuracy": 0.15592513233423233,
"num_tokens": 12944078.0,
"step": 7460
},
{
"entropy": 5.635997152328491,
"epoch": 0.5808208519743241,
"grad_norm": 1.6015625,
"learning_rate": 0.0004971530465939005,
"loss": 5.548,
"mean_token_accuracy": 0.1559380941092968,
"num_tokens": 12952800.0,
"step": 7465
},
{
"entropy": 5.724119472503662,
"epoch": 0.5812098813460417,
"grad_norm": 1.515625,
"learning_rate": 0.0004971486498986825,
"loss": 5.6344,
"mean_token_accuracy": 0.15321501046419145,
"num_tokens": 12962402.0,
"step": 7470
},
{
"entropy": 5.621135282516479,
"epoch": 0.5815989107177592,
"grad_norm": 1.296875,
"learning_rate": 0.0004971442498327124,
"loss": 5.4026,
"mean_token_accuracy": 0.16943953484296798,
"num_tokens": 12971116.0,
"step": 7475
},
{
"entropy": 5.5255170345306395,
"epoch": 0.5819879400894767,
"grad_norm": 1.4453125,
"learning_rate": 0.0004971398463960568,
"loss": 5.5058,
"mean_token_accuracy": 0.1556493379175663,
"num_tokens": 12979570.0,
"step": 7480
},
{
"entropy": 5.673506450653076,
"epoch": 0.5823769694611943,
"grad_norm": 1.578125,
"learning_rate": 0.0004971354395887827,
"loss": 5.5064,
"mean_token_accuracy": 0.16239574402570725,
"num_tokens": 12987942.0,
"step": 7485
},
{
"entropy": 5.678384733200073,
"epoch": 0.5827659988329119,
"grad_norm": 1.6328125,
"learning_rate": 0.0004971310294109569,
"loss": 5.5726,
"mean_token_accuracy": 0.15900402516126633,
"num_tokens": 12996499.0,
"step": 7490
},
{
"entropy": 5.596940183639527,
"epoch": 0.5831550282046295,
"grad_norm": 1.2734375,
"learning_rate": 0.0004971266158626461,
"loss": 5.5215,
"mean_token_accuracy": 0.16408492773771285,
"num_tokens": 13005622.0,
"step": 7495
},
{
"entropy": 5.666322374343872,
"epoch": 0.583544057576347,
"grad_norm": 1.3984375,
"learning_rate": 0.0004971221989439177,
"loss": 5.5131,
"mean_token_accuracy": 0.1555515393614769,
"num_tokens": 13014417.0,
"step": 7500
},
{
"entropy": 5.591904401779175,
"epoch": 0.5839330869480646,
"grad_norm": 1.7734375,
"learning_rate": 0.0004971177786548384,
"loss": 5.4505,
"mean_token_accuracy": 0.1593570254743099,
"num_tokens": 13022887.0,
"step": 7505
},
{
"entropy": 5.598121070861817,
"epoch": 0.5843221163197821,
"grad_norm": 1.3203125,
"learning_rate": 0.0004971133549954753,
"loss": 5.5523,
"mean_token_accuracy": 0.15993799567222594,
"num_tokens": 13031676.0,
"step": 7510
},
{
"entropy": 5.591425514221191,
"epoch": 0.5847111456914997,
"grad_norm": 1.40625,
"learning_rate": 0.0004971089279658954,
"loss": 5.5058,
"mean_token_accuracy": 0.1619194507598877,
"num_tokens": 13040489.0,
"step": 7515
},
{
"entropy": 5.681172895431518,
"epoch": 0.5851001750632173,
"grad_norm": 1.3515625,
"learning_rate": 0.0004971044975661663,
"loss": 5.4316,
"mean_token_accuracy": 0.16690255403518678,
"num_tokens": 13048280.0,
"step": 7520
},
{
"entropy": 5.5702001571655275,
"epoch": 0.5854892044349348,
"grad_norm": 1.484375,
"learning_rate": 0.0004971000637963549,
"loss": 5.4192,
"mean_token_accuracy": 0.1744627207517624,
"num_tokens": 13056471.0,
"step": 7525
},
{
"entropy": 5.584231662750244,
"epoch": 0.5858782338066524,
"grad_norm": 1.421875,
"learning_rate": 0.0004970956266565285,
"loss": 5.6039,
"mean_token_accuracy": 0.15952491909265518,
"num_tokens": 13065289.0,
"step": 7530
},
{
"entropy": 5.666153907775879,
"epoch": 0.58626726317837,
"grad_norm": 1.703125,
"learning_rate": 0.0004970911861467544,
"loss": 5.6199,
"mean_token_accuracy": 0.15906012058258057,
"num_tokens": 13073648.0,
"step": 7535
},
{
"entropy": 5.713760423660278,
"epoch": 0.5866562925500876,
"grad_norm": 1.4140625,
"learning_rate": 0.0004970867422671002,
"loss": 5.6328,
"mean_token_accuracy": 0.15964424088597298,
"num_tokens": 13082205.0,
"step": 7540
},
{
"entropy": 5.688720226287842,
"epoch": 0.5870453219218051,
"grad_norm": 1.640625,
"learning_rate": 0.000497082295017633,
"loss": 5.5039,
"mean_token_accuracy": 0.15762317776679993,
"num_tokens": 13090970.0,
"step": 7545
},
{
"entropy": 5.660140466690064,
"epoch": 0.5874343512935226,
"grad_norm": 1.640625,
"learning_rate": 0.0004970778443984206,
"loss": 5.408,
"mean_token_accuracy": 0.16850501745939256,
"num_tokens": 13098290.0,
"step": 7550
},
{
"entropy": 5.521188640594483,
"epoch": 0.5878233806652402,
"grad_norm": 1.2890625,
"learning_rate": 0.0004970733904095303,
"loss": 5.3654,
"mean_token_accuracy": 0.1694828599691391,
"num_tokens": 13106971.0,
"step": 7555
},
{
"entropy": 5.6553733348846436,
"epoch": 0.5882124100369578,
"grad_norm": 1.3515625,
"learning_rate": 0.0004970689330510298,
"loss": 5.6533,
"mean_token_accuracy": 0.15376500338315963,
"num_tokens": 13115742.0,
"step": 7560
},
{
"entropy": 5.676582717895508,
"epoch": 0.5886014394086754,
"grad_norm": 1.5625,
"learning_rate": 0.0004970644723229868,
"loss": 5.6128,
"mean_token_accuracy": 0.15956298857927323,
"num_tokens": 13124192.0,
"step": 7565
},
{
"entropy": 5.625477743148804,
"epoch": 0.588990468780393,
"grad_norm": 1.375,
"learning_rate": 0.0004970600082254687,
"loss": 5.4197,
"mean_token_accuracy": 0.16239960491657257,
"num_tokens": 13133510.0,
"step": 7570
},
{
"entropy": 5.592658853530883,
"epoch": 0.5893794981521104,
"grad_norm": 1.546875,
"learning_rate": 0.0004970555407585436,
"loss": 5.5552,
"mean_token_accuracy": 0.1599623069167137,
"num_tokens": 13141704.0,
"step": 7575
},
{
"entropy": 5.518377590179443,
"epoch": 0.589768527523828,
"grad_norm": 1.5625,
"learning_rate": 0.000497051069922279,
"loss": 5.4366,
"mean_token_accuracy": 0.15510185062885284,
"num_tokens": 13150559.0,
"step": 7580
},
{
"entropy": 5.581053924560547,
"epoch": 0.5901575568955456,
"grad_norm": 1.5078125,
"learning_rate": 0.0004970465957167429,
"loss": 5.4834,
"mean_token_accuracy": 0.15990304946899414,
"num_tokens": 13159428.0,
"step": 7585
},
{
"entropy": 5.644467401504516,
"epoch": 0.5905465862672632,
"grad_norm": 1.5390625,
"learning_rate": 0.000497042118142003,
"loss": 5.6244,
"mean_token_accuracy": 0.15349241942167283,
"num_tokens": 13168299.0,
"step": 7590
},
{
"entropy": 5.6267036437988285,
"epoch": 0.5909356156389808,
"grad_norm": 1.359375,
"learning_rate": 0.0004970376371981275,
"loss": 5.4908,
"mean_token_accuracy": 0.15986540913581848,
"num_tokens": 13176906.0,
"step": 7595
},
{
"entropy": 5.735187244415283,
"epoch": 0.5913246450106983,
"grad_norm": 1.390625,
"learning_rate": 0.0004970331528851842,
"loss": 5.5883,
"mean_token_accuracy": 0.15484750866889954,
"num_tokens": 13186180.0,
"step": 7600
},
{
"entropy": 5.550314044952392,
"epoch": 0.5917136743824158,
"grad_norm": 1.4453125,
"learning_rate": 0.0004970286652032412,
"loss": 5.4165,
"mean_token_accuracy": 0.16556940227746964,
"num_tokens": 13195111.0,
"step": 7605
},
{
"entropy": 5.586171960830688,
"epoch": 0.5921027037541334,
"grad_norm": 1.3984375,
"learning_rate": 0.0004970241741523667,
"loss": 5.4881,
"mean_token_accuracy": 0.15955300033092498,
"num_tokens": 13204170.0,
"step": 7610
},
{
"entropy": 5.677660846710205,
"epoch": 0.592491733125851,
"grad_norm": 1.546875,
"learning_rate": 0.0004970196797326286,
"loss": 5.5105,
"mean_token_accuracy": 0.15856559723615646,
"num_tokens": 13213479.0,
"step": 7615
},
{
"entropy": 5.639448738098144,
"epoch": 0.5928807624975686,
"grad_norm": 1.3359375,
"learning_rate": 0.0004970151819440955,
"loss": 5.5509,
"mean_token_accuracy": 0.15828660875558853,
"num_tokens": 13222182.0,
"step": 7620
},
{
"entropy": 5.531005382537842,
"epoch": 0.5932697918692861,
"grad_norm": 1.2578125,
"learning_rate": 0.0004970106807868351,
"loss": 5.4032,
"mean_token_accuracy": 0.16099072992801666,
"num_tokens": 13230681.0,
"step": 7625
},
{
"entropy": 5.700342130661011,
"epoch": 0.5936588212410037,
"grad_norm": 1.375,
"learning_rate": 0.000497006176260916,
"loss": 5.5912,
"mean_token_accuracy": 0.1595593973994255,
"num_tokens": 13239254.0,
"step": 7630
},
{
"entropy": 5.617882537841797,
"epoch": 0.5940478506127213,
"grad_norm": 1.390625,
"learning_rate": 0.0004970016683664068,
"loss": 5.4469,
"mean_token_accuracy": 0.16579967588186265,
"num_tokens": 13248317.0,
"step": 7635
},
{
"entropy": 5.654610681533813,
"epoch": 0.5944368799844388,
"grad_norm": 1.921875,
"learning_rate": 0.0004969971571033754,
"loss": 5.5053,
"mean_token_accuracy": 0.15919456630945206,
"num_tokens": 13256977.0,
"step": 7640
},
{
"entropy": 5.615067386627198,
"epoch": 0.5948259093561564,
"grad_norm": 1.5,
"learning_rate": 0.0004969926424718906,
"loss": 5.5841,
"mean_token_accuracy": 0.15947894304990767,
"num_tokens": 13266012.0,
"step": 7645
},
{
"entropy": 5.707496690750122,
"epoch": 0.5952149387278739,
"grad_norm": 1.3984375,
"learning_rate": 0.0004969881244720207,
"loss": 5.6046,
"mean_token_accuracy": 0.1607809454202652,
"num_tokens": 13275248.0,
"step": 7650
},
{
"entropy": 5.637995338439941,
"epoch": 0.5956039680995915,
"grad_norm": 1.390625,
"learning_rate": 0.0004969836031038343,
"loss": 5.4403,
"mean_token_accuracy": 0.1559646725654602,
"num_tokens": 13284235.0,
"step": 7655
},
{
"entropy": 5.573863458633423,
"epoch": 0.5959929974713091,
"grad_norm": 1.390625,
"learning_rate": 0.0004969790783674002,
"loss": 5.5246,
"mean_token_accuracy": 0.1523687422275543,
"num_tokens": 13292423.0,
"step": 7660
},
{
"entropy": 5.6371039867401125,
"epoch": 0.5963820268430267,
"grad_norm": 1.328125,
"learning_rate": 0.0004969745502627868,
"loss": 5.5628,
"mean_token_accuracy": 0.15199742317199708,
"num_tokens": 13300936.0,
"step": 7665
},
{
"entropy": 5.552597761154175,
"epoch": 0.5967710562147442,
"grad_norm": 1.3125,
"learning_rate": 0.0004969700187900629,
"loss": 5.3773,
"mean_token_accuracy": 0.17030118703842162,
"num_tokens": 13308881.0,
"step": 7670
},
{
"entropy": 5.61635012626648,
"epoch": 0.5971600855864618,
"grad_norm": 1.375,
"learning_rate": 0.0004969654839492973,
"loss": 5.5448,
"mean_token_accuracy": 0.15784391164779663,
"num_tokens": 13317864.0,
"step": 7675
},
{
"entropy": 5.657650375366211,
"epoch": 0.5975491149581793,
"grad_norm": 1.4453125,
"learning_rate": 0.0004969609457405588,
"loss": 5.5087,
"mean_token_accuracy": 0.15694525986909866,
"num_tokens": 13326990.0,
"step": 7680
},
{
"entropy": 5.748386859893799,
"epoch": 0.5979381443298969,
"grad_norm": 1.453125,
"learning_rate": 0.0004969564041639163,
"loss": 5.6993,
"mean_token_accuracy": 0.14893480986356736,
"num_tokens": 13336437.0,
"step": 7685
},
{
"entropy": 5.663937950134278,
"epoch": 0.5983271737016145,
"grad_norm": 1.3828125,
"learning_rate": 0.0004969518592194386,
"loss": 5.5151,
"mean_token_accuracy": 0.15797433406114578,
"num_tokens": 13345368.0,
"step": 7690
},
{
"entropy": 5.5918773174285885,
"epoch": 0.5987162030733321,
"grad_norm": 1.2734375,
"learning_rate": 0.0004969473109071946,
"loss": 5.5545,
"mean_token_accuracy": 0.1600046530365944,
"num_tokens": 13354171.0,
"step": 7695
},
{
"entropy": 5.682364416122437,
"epoch": 0.5991052324450497,
"grad_norm": 1.3359375,
"learning_rate": 0.0004969427592272535,
"loss": 5.5552,
"mean_token_accuracy": 0.15920861437916756,
"num_tokens": 13362912.0,
"step": 7700
},
{
"entropy": 5.603416681289673,
"epoch": 0.5994942618167671,
"grad_norm": 1.3984375,
"learning_rate": 0.0004969382041796843,
"loss": 5.4395,
"mean_token_accuracy": 0.1599783182144165,
"num_tokens": 13370913.0,
"step": 7705
},
{
"entropy": 5.660623693466187,
"epoch": 0.5998832911884847,
"grad_norm": 1.953125,
"learning_rate": 0.000496933645764556,
"loss": 5.6038,
"mean_token_accuracy": 0.1545683339238167,
"num_tokens": 13379557.0,
"step": 7710
},
{
"entropy": 5.658495330810547,
"epoch": 0.6002723205602023,
"grad_norm": 1.4140625,
"learning_rate": 0.0004969290839819381,
"loss": 5.511,
"mean_token_accuracy": 0.1654706969857216,
"num_tokens": 13388129.0,
"step": 7715
},
{
"entropy": 5.644501495361328,
"epoch": 0.6006613499319199,
"grad_norm": 1.5078125,
"learning_rate": 0.0004969245188318994,
"loss": 5.6286,
"mean_token_accuracy": 0.15689315497875214,
"num_tokens": 13396713.0,
"step": 7720
},
{
"entropy": 5.6918860912323,
"epoch": 0.6010503793036375,
"grad_norm": 1.3828125,
"learning_rate": 0.0004969199503145093,
"loss": 5.5562,
"mean_token_accuracy": 0.16171080321073533,
"num_tokens": 13406171.0,
"step": 7725
},
{
"entropy": 5.5995087146759035,
"epoch": 0.601439408675355,
"grad_norm": 1.3828125,
"learning_rate": 0.0004969153784298374,
"loss": 5.4815,
"mean_token_accuracy": 0.1628172367811203,
"num_tokens": 13415107.0,
"step": 7730
},
{
"entropy": 5.590442943572998,
"epoch": 0.6018284380470725,
"grad_norm": 1.546875,
"learning_rate": 0.0004969108031779527,
"loss": 5.5996,
"mean_token_accuracy": 0.16061357855796815,
"num_tokens": 13423533.0,
"step": 7735
},
{
"entropy": 5.716766357421875,
"epoch": 0.6022174674187901,
"grad_norm": 1.5390625,
"learning_rate": 0.0004969062245589247,
"loss": 5.565,
"mean_token_accuracy": 0.15467922687530516,
"num_tokens": 13432798.0,
"step": 7740
},
{
"entropy": 5.691170167922974,
"epoch": 0.6026064967905077,
"grad_norm": 1.453125,
"learning_rate": 0.0004969016425728231,
"loss": 5.5097,
"mean_token_accuracy": 0.16404221951961517,
"num_tokens": 13441123.0,
"step": 7745
},
{
"entropy": 5.564320659637451,
"epoch": 0.6029955261622253,
"grad_norm": 1.390625,
"learning_rate": 0.0004968970572197172,
"loss": 5.4399,
"mean_token_accuracy": 0.16656853556632994,
"num_tokens": 13449548.0,
"step": 7750
},
{
"entropy": 5.631894779205322,
"epoch": 0.6033845555339428,
"grad_norm": 1.46875,
"learning_rate": 0.0004968924684996765,
"loss": 5.5268,
"mean_token_accuracy": 0.15824869126081467,
"num_tokens": 13458022.0,
"step": 7755
},
{
"entropy": 5.609321880340576,
"epoch": 0.6037735849056604,
"grad_norm": 1.453125,
"learning_rate": 0.000496887876412771,
"loss": 5.4797,
"mean_token_accuracy": 0.16207483112812043,
"num_tokens": 13466871.0,
"step": 7760
},
{
"entropy": 5.600290250778198,
"epoch": 0.6041626142773779,
"grad_norm": 1.3671875,
"learning_rate": 0.0004968832809590701,
"loss": 5.5171,
"mean_token_accuracy": 0.16046020835638047,
"num_tokens": 13475659.0,
"step": 7765
},
{
"entropy": 5.656421041488647,
"epoch": 0.6045516436490955,
"grad_norm": 1.453125,
"learning_rate": 0.0004968786821386435,
"loss": 5.5303,
"mean_token_accuracy": 0.16244256794452666,
"num_tokens": 13484146.0,
"step": 7770
},
{
"entropy": 5.684045696258545,
"epoch": 0.6049406730208131,
"grad_norm": 1.2734375,
"learning_rate": 0.0004968740799515611,
"loss": 5.5367,
"mean_token_accuracy": 0.16123472899198532,
"num_tokens": 13492831.0,
"step": 7775
},
{
"entropy": 5.642293262481689,
"epoch": 0.6053297023925306,
"grad_norm": 1.4296875,
"learning_rate": 0.0004968694743978926,
"loss": 5.503,
"mean_token_accuracy": 0.16224339455366135,
"num_tokens": 13501640.0,
"step": 7780
},
{
"entropy": 5.707996892929077,
"epoch": 0.6057187317642482,
"grad_norm": 1.3359375,
"learning_rate": 0.0004968648654777081,
"loss": 5.5145,
"mean_token_accuracy": 0.16158525943756102,
"num_tokens": 13509816.0,
"step": 7785
},
{
"entropy": 5.663367700576782,
"epoch": 0.6061077611359658,
"grad_norm": 1.453125,
"learning_rate": 0.0004968602531910773,
"loss": 5.5868,
"mean_token_accuracy": 0.15522609949111937,
"num_tokens": 13518384.0,
"step": 7790
},
{
"entropy": 5.551813411712646,
"epoch": 0.6064967905076833,
"grad_norm": 1.4453125,
"learning_rate": 0.0004968556375380704,
"loss": 5.4464,
"mean_token_accuracy": 0.1610630139708519,
"num_tokens": 13527115.0,
"step": 7795
},
{
"entropy": 5.6546814918518065,
"epoch": 0.6068858198794009,
"grad_norm": 1.359375,
"learning_rate": 0.0004968510185187571,
"loss": 5.6641,
"mean_token_accuracy": 0.1552654519677162,
"num_tokens": 13535743.0,
"step": 7800
},
{
"entropy": 5.596662521362305,
"epoch": 0.6072748492511184,
"grad_norm": 1.421875,
"learning_rate": 0.000496846396133208,
"loss": 5.4717,
"mean_token_accuracy": 0.16270544081926347,
"num_tokens": 13544167.0,
"step": 7805
},
{
"entropy": 5.643411493301391,
"epoch": 0.607663878622836,
"grad_norm": 1.3828125,
"learning_rate": 0.0004968417703814928,
"loss": 5.5246,
"mean_token_accuracy": 0.15955984592437744,
"num_tokens": 13552291.0,
"step": 7810
},
{
"entropy": 5.669795370101928,
"epoch": 0.6080529079945536,
"grad_norm": 1.421875,
"learning_rate": 0.0004968371412636818,
"loss": 5.5821,
"mean_token_accuracy": 0.15759380161762238,
"num_tokens": 13561185.0,
"step": 7815
},
{
"entropy": 5.650271654129028,
"epoch": 0.6084419373662712,
"grad_norm": 1.3671875,
"learning_rate": 0.0004968325087798453,
"loss": 5.4608,
"mean_token_accuracy": 0.1675198793411255,
"num_tokens": 13569815.0,
"step": 7820
},
{
"entropy": 5.66109619140625,
"epoch": 0.6088309667379888,
"grad_norm": 1.3671875,
"learning_rate": 0.0004968278729300536,
"loss": 5.5835,
"mean_token_accuracy": 0.15778862088918685,
"num_tokens": 13578882.0,
"step": 7825
},
{
"entropy": 5.557735729217529,
"epoch": 0.6092199961097062,
"grad_norm": 1.546875,
"learning_rate": 0.0004968232337143769,
"loss": 5.4403,
"mean_token_accuracy": 0.16673890501260757,
"num_tokens": 13587479.0,
"step": 7830
},
{
"entropy": 5.47802152633667,
"epoch": 0.6096090254814238,
"grad_norm": 1.3125,
"learning_rate": 0.0004968185911328858,
"loss": 5.3486,
"mean_token_accuracy": 0.16249027252197265,
"num_tokens": 13596049.0,
"step": 7835
},
{
"entropy": 5.684981775283814,
"epoch": 0.6099980548531414,
"grad_norm": 1.2265625,
"learning_rate": 0.0004968139451856506,
"loss": 5.6045,
"mean_token_accuracy": 0.1551737293601036,
"num_tokens": 13604766.0,
"step": 7840
},
{
"entropy": 5.684131145477295,
"epoch": 0.610387084224859,
"grad_norm": 1.2890625,
"learning_rate": 0.0004968092958727419,
"loss": 5.5887,
"mean_token_accuracy": 0.1608073055744171,
"num_tokens": 13614051.0,
"step": 7845
},
{
"entropy": 5.571271324157715,
"epoch": 0.6107761135965766,
"grad_norm": 1.3046875,
"learning_rate": 0.00049680464319423,
"loss": 5.5338,
"mean_token_accuracy": 0.15778470188379287,
"num_tokens": 13623100.0,
"step": 7850
},
{
"entropy": 5.699359178543091,
"epoch": 0.6111651429682942,
"grad_norm": 1.34375,
"learning_rate": 0.0004967999871501858,
"loss": 5.552,
"mean_token_accuracy": 0.15810873508453369,
"num_tokens": 13631732.0,
"step": 7855
},
{
"entropy": 5.61888165473938,
"epoch": 0.6115541723400116,
"grad_norm": 1.703125,
"learning_rate": 0.0004967953277406798,
"loss": 5.5358,
"mean_token_accuracy": 0.16011350452899933,
"num_tokens": 13640355.0,
"step": 7860
},
{
"entropy": 5.630750751495361,
"epoch": 0.6119432017117292,
"grad_norm": 1.5703125,
"learning_rate": 0.0004967906649657828,
"loss": 5.6121,
"mean_token_accuracy": 0.16256681680679322,
"num_tokens": 13649108.0,
"step": 7865
},
{
"entropy": 5.613716268539429,
"epoch": 0.6123322310834468,
"grad_norm": 1.46875,
"learning_rate": 0.0004967859988255655,
"loss": 5.4381,
"mean_token_accuracy": 0.17376051247119903,
"num_tokens": 13656765.0,
"step": 7870
},
{
"entropy": 5.616294431686401,
"epoch": 0.6127212604551644,
"grad_norm": 1.5390625,
"learning_rate": 0.0004967813293200985,
"loss": 5.529,
"mean_token_accuracy": 0.15748461037874223,
"num_tokens": 13664692.0,
"step": 7875
},
{
"entropy": 5.500170755386352,
"epoch": 0.613110289826882,
"grad_norm": 1.234375,
"learning_rate": 0.0004967766564494529,
"loss": 5.3894,
"mean_token_accuracy": 0.16902236491441727,
"num_tokens": 13672913.0,
"step": 7880
},
{
"entropy": 5.559022665023804,
"epoch": 0.6134993191985995,
"grad_norm": 1.359375,
"learning_rate": 0.0004967719802136996,
"loss": 5.3854,
"mean_token_accuracy": 0.16822486221790314,
"num_tokens": 13682427.0,
"step": 7885
},
{
"entropy": 5.612292194366455,
"epoch": 0.613888348570317,
"grad_norm": 1.390625,
"learning_rate": 0.0004967673006129094,
"loss": 5.4823,
"mean_token_accuracy": 0.16599115133285522,
"num_tokens": 13690815.0,
"step": 7890
},
{
"entropy": 5.672644567489624,
"epoch": 0.6142773779420346,
"grad_norm": 1.3671875,
"learning_rate": 0.0004967626176471536,
"loss": 5.6092,
"mean_token_accuracy": 0.15415968373417854,
"num_tokens": 13699522.0,
"step": 7895
},
{
"entropy": 5.705391597747803,
"epoch": 0.6146664073137522,
"grad_norm": 1.3671875,
"learning_rate": 0.0004967579313165028,
"loss": 5.618,
"mean_token_accuracy": 0.1559215635061264,
"num_tokens": 13708075.0,
"step": 7900
},
{
"entropy": 5.673476839065552,
"epoch": 0.6150554366854698,
"grad_norm": 1.2578125,
"learning_rate": 0.0004967532416210284,
"loss": 5.4947,
"mean_token_accuracy": 0.16200094670057297,
"num_tokens": 13716397.0,
"step": 7905
},
{
"entropy": 5.5525562286376955,
"epoch": 0.6154444660571873,
"grad_norm": 1.296875,
"learning_rate": 0.0004967485485608016,
"loss": 5.5068,
"mean_token_accuracy": 0.16395258158445358,
"num_tokens": 13724724.0,
"step": 7910
},
{
"entropy": 5.668342161178589,
"epoch": 0.6158334954289049,
"grad_norm": 1.296875,
"learning_rate": 0.0004967438521358934,
"loss": 5.4794,
"mean_token_accuracy": 0.16554345041513444,
"num_tokens": 13733389.0,
"step": 7915
},
{
"entropy": 5.5953020572662355,
"epoch": 0.6162225248006225,
"grad_norm": 1.4453125,
"learning_rate": 0.0004967391523463754,
"loss": 5.4385,
"mean_token_accuracy": 0.1632681056857109,
"num_tokens": 13742569.0,
"step": 7920
},
{
"entropy": 5.581040382385254,
"epoch": 0.61661155417234,
"grad_norm": 1.5078125,
"learning_rate": 0.0004967344491923185,
"loss": 5.4168,
"mean_token_accuracy": 0.16416201442480088,
"num_tokens": 13750975.0,
"step": 7925
},
{
"entropy": 5.57597918510437,
"epoch": 0.6170005835440576,
"grad_norm": 1.390625,
"learning_rate": 0.0004967297426737943,
"loss": 5.4778,
"mean_token_accuracy": 0.16318628638982774,
"num_tokens": 13758782.0,
"step": 7930
},
{
"entropy": 5.630909013748169,
"epoch": 0.6173896129157751,
"grad_norm": 1.4296875,
"learning_rate": 0.0004967250327908742,
"loss": 5.5922,
"mean_token_accuracy": 0.15122650414705277,
"num_tokens": 13767321.0,
"step": 7935
},
{
"entropy": 5.693921947479248,
"epoch": 0.6177786422874927,
"grad_norm": 1.421875,
"learning_rate": 0.0004967203195436297,
"loss": 5.5725,
"mean_token_accuracy": 0.16585950404405594,
"num_tokens": 13775992.0,
"step": 7940
},
{
"entropy": 5.578035116195679,
"epoch": 0.6181676716592103,
"grad_norm": 1.359375,
"learning_rate": 0.0004967156029321322,
"loss": 5.4309,
"mean_token_accuracy": 0.16884675174951552,
"num_tokens": 13784358.0,
"step": 7945
},
{
"entropy": 5.484552240371704,
"epoch": 0.6185567010309279,
"grad_norm": 1.4296875,
"learning_rate": 0.0004967108829564532,
"loss": 5.4268,
"mean_token_accuracy": 0.16732152700424194,
"num_tokens": 13792834.0,
"step": 7950
},
{
"entropy": 5.632082462310791,
"epoch": 0.6189457304026454,
"grad_norm": 1.296875,
"learning_rate": 0.0004967061596166646,
"loss": 5.5165,
"mean_token_accuracy": 0.15893340855836868,
"num_tokens": 13801414.0,
"step": 7955
},
{
"entropy": 5.6532261848449705,
"epoch": 0.6193347597743629,
"grad_norm": 1.3515625,
"learning_rate": 0.0004967014329128378,
"loss": 5.5581,
"mean_token_accuracy": 0.16025267094373702,
"num_tokens": 13810459.0,
"step": 7960
},
{
"entropy": 5.592121505737305,
"epoch": 0.6197237891460805,
"grad_norm": 1.484375,
"learning_rate": 0.0004966967028450448,
"loss": 5.3791,
"mean_token_accuracy": 0.16667962819337845,
"num_tokens": 13818460.0,
"step": 7965
},
{
"entropy": 5.554238891601562,
"epoch": 0.6201128185177981,
"grad_norm": 1.34375,
"learning_rate": 0.000496691969413357,
"loss": 5.507,
"mean_token_accuracy": 0.16117742657661438,
"num_tokens": 13827693.0,
"step": 7970
},
{
"entropy": 5.683591508865357,
"epoch": 0.6205018478895157,
"grad_norm": 1.4296875,
"learning_rate": 0.0004966872326178466,
"loss": 5.5158,
"mean_token_accuracy": 0.15253090113401413,
"num_tokens": 13836506.0,
"step": 7975
},
{
"entropy": 5.642145490646362,
"epoch": 0.6208908772612333,
"grad_norm": 1.453125,
"learning_rate": 0.0004966824924585851,
"loss": 5.5238,
"mean_token_accuracy": 0.15704460814595222,
"num_tokens": 13844977.0,
"step": 7980
},
{
"entropy": 5.580069446563721,
"epoch": 0.6212799066329507,
"grad_norm": 1.3515625,
"learning_rate": 0.0004966777489356448,
"loss": 5.5077,
"mean_token_accuracy": 0.15980817675590514,
"num_tokens": 13854514.0,
"step": 7985
},
{
"entropy": 5.645097970962524,
"epoch": 0.6216689360046683,
"grad_norm": 1.3125,
"learning_rate": 0.0004966730020490975,
"loss": 5.5072,
"mean_token_accuracy": 0.1550505816936493,
"num_tokens": 13863037.0,
"step": 7990
},
{
"entropy": 5.599818134307862,
"epoch": 0.6220579653763859,
"grad_norm": 1.4296875,
"learning_rate": 0.000496668251799015,
"loss": 5.4687,
"mean_token_accuracy": 0.1692013517022133,
"num_tokens": 13871720.0,
"step": 7995
},
{
"entropy": 5.636774635314941,
"epoch": 0.6224469947481035,
"grad_norm": 1.5546875,
"learning_rate": 0.0004966634981854699,
"loss": 5.5026,
"mean_token_accuracy": 0.16675081104040146,
"num_tokens": 13880214.0,
"step": 8000
},
{
"entropy": 5.616530132293701,
"epoch": 0.6228360241198211,
"grad_norm": 1.3125,
"learning_rate": 0.0004966587412085339,
"loss": 5.5975,
"mean_token_accuracy": 0.152876765280962,
"num_tokens": 13888784.0,
"step": 8005
},
{
"entropy": 5.650445365905762,
"epoch": 0.6232250534915386,
"grad_norm": 1.4296875,
"learning_rate": 0.0004966539808682793,
"loss": 5.5021,
"mean_token_accuracy": 0.16107800453901291,
"num_tokens": 13897466.0,
"step": 8010
},
{
"entropy": 5.589649438858032,
"epoch": 0.6236140828632561,
"grad_norm": 1.46875,
"learning_rate": 0.0004966492171647783,
"loss": 5.5146,
"mean_token_accuracy": 0.15655134320259095,
"num_tokens": 13906556.0,
"step": 8015
},
{
"entropy": 5.555747985839844,
"epoch": 0.6240031122349737,
"grad_norm": 1.4140625,
"learning_rate": 0.0004966444500981032,
"loss": 5.4576,
"mean_token_accuracy": 0.16158139556646348,
"num_tokens": 13915473.0,
"step": 8020
},
{
"entropy": 5.653251791000367,
"epoch": 0.6243921416066913,
"grad_norm": 1.5546875,
"learning_rate": 0.0004966396796683265,
"loss": 5.6261,
"mean_token_accuracy": 0.15274745523929595,
"num_tokens": 13924775.0,
"step": 8025
},
{
"entropy": 5.692790603637695,
"epoch": 0.6247811709784089,
"grad_norm": 1.421875,
"learning_rate": 0.0004966349058755204,
"loss": 5.5036,
"mean_token_accuracy": 0.15476525872945784,
"num_tokens": 13933992.0,
"step": 8030
},
{
"entropy": 5.731003522872925,
"epoch": 0.6251702003501264,
"grad_norm": 1.5078125,
"learning_rate": 0.0004966301287197573,
"loss": 5.6728,
"mean_token_accuracy": 0.15809391736984252,
"num_tokens": 13942877.0,
"step": 8035
},
{
"entropy": 5.636298751831054,
"epoch": 0.625559229721844,
"grad_norm": 1.2734375,
"learning_rate": 0.0004966253482011098,
"loss": 5.5205,
"mean_token_accuracy": 0.16148761957883834,
"num_tokens": 13952003.0,
"step": 8040
},
{
"entropy": 5.6766040325164795,
"epoch": 0.6259482590935616,
"grad_norm": 1.4609375,
"learning_rate": 0.0004966205643196503,
"loss": 5.4583,
"mean_token_accuracy": 0.16635134518146516,
"num_tokens": 13959801.0,
"step": 8045
},
{
"entropy": 5.51854133605957,
"epoch": 0.6263372884652791,
"grad_norm": 1.4375,
"learning_rate": 0.0004966157770754516,
"loss": 5.4561,
"mean_token_accuracy": 0.15555119663476943,
"num_tokens": 13968535.0,
"step": 8050
},
{
"entropy": 5.647353887557983,
"epoch": 0.6267263178369967,
"grad_norm": 1.4765625,
"learning_rate": 0.000496610986468586,
"loss": 5.445,
"mean_token_accuracy": 0.16241587996482848,
"num_tokens": 13977079.0,
"step": 8055
},
{
"entropy": 5.602263879776001,
"epoch": 0.6271153472087143,
"grad_norm": 1.3515625,
"learning_rate": 0.0004966061924991267,
"loss": 5.4639,
"mean_token_accuracy": 0.17081388980150222,
"num_tokens": 13985118.0,
"step": 8060
},
{
"entropy": 5.552741384506225,
"epoch": 0.6275043765804318,
"grad_norm": 1.3203125,
"learning_rate": 0.000496601395167146,
"loss": 5.4125,
"mean_token_accuracy": 0.16479296535253524,
"num_tokens": 13994061.0,
"step": 8065
},
{
"entropy": 5.61691837310791,
"epoch": 0.6278934059521494,
"grad_norm": 1.2578125,
"learning_rate": 0.0004965965944727168,
"loss": 5.5592,
"mean_token_accuracy": 0.1568474255502224,
"num_tokens": 14002879.0,
"step": 8070
},
{
"entropy": 5.667020177841186,
"epoch": 0.628282435323867,
"grad_norm": 1.5234375,
"learning_rate": 0.0004965917904159121,
"loss": 5.5355,
"mean_token_accuracy": 0.1621012195944786,
"num_tokens": 14011570.0,
"step": 8075
},
{
"entropy": 5.573873233795166,
"epoch": 0.6286714646955845,
"grad_norm": 1.3046875,
"learning_rate": 0.0004965869829968046,
"loss": 5.3767,
"mean_token_accuracy": 0.16462417393922807,
"num_tokens": 14019282.0,
"step": 8080
},
{
"entropy": 5.547157669067383,
"epoch": 0.6290604940673021,
"grad_norm": 1.6015625,
"learning_rate": 0.0004965821722154674,
"loss": 5.3898,
"mean_token_accuracy": 0.1729328453540802,
"num_tokens": 14027774.0,
"step": 8085
},
{
"entropy": 5.598388719558716,
"epoch": 0.6294495234390196,
"grad_norm": 1.421875,
"learning_rate": 0.0004965773580719734,
"loss": 5.4058,
"mean_token_accuracy": 0.1658223569393158,
"num_tokens": 14036754.0,
"step": 8090
},
{
"entropy": 5.665834665298462,
"epoch": 0.6298385528107372,
"grad_norm": 1.328125,
"learning_rate": 0.0004965725405663957,
"loss": 5.5791,
"mean_token_accuracy": 0.16007334589958191,
"num_tokens": 14045714.0,
"step": 8095
},
{
"entropy": 5.676879072189331,
"epoch": 0.6302275821824548,
"grad_norm": 2.078125,
"learning_rate": 0.0004965677196988073,
"loss": 5.5926,
"mean_token_accuracy": 0.1579957962036133,
"num_tokens": 14054973.0,
"step": 8100
},
{
"entropy": 5.616937971115112,
"epoch": 0.6306166115541724,
"grad_norm": 1.40625,
"learning_rate": 0.0004965628954692815,
"loss": 5.4947,
"mean_token_accuracy": 0.1623809590935707,
"num_tokens": 14064037.0,
"step": 8105
},
{
"entropy": 5.661908102035523,
"epoch": 0.63100564092589,
"grad_norm": 1.484375,
"learning_rate": 0.0004965580678778913,
"loss": 5.5264,
"mean_token_accuracy": 0.16121605634689332,
"num_tokens": 14072269.0,
"step": 8110
},
{
"entropy": 5.621046781539917,
"epoch": 0.6313946702976074,
"grad_norm": 1.484375,
"learning_rate": 0.0004965532369247102,
"loss": 5.4633,
"mean_token_accuracy": 0.16086421832442283,
"num_tokens": 14080595.0,
"step": 8115
},
{
"entropy": 5.58659930229187,
"epoch": 0.631783699669325,
"grad_norm": 1.40625,
"learning_rate": 0.0004965484026098111,
"loss": 5.4495,
"mean_token_accuracy": 0.1654338777065277,
"num_tokens": 14089159.0,
"step": 8120
},
{
"entropy": 5.630824947357178,
"epoch": 0.6321727290410426,
"grad_norm": 1.484375,
"learning_rate": 0.0004965435649332679,
"loss": 5.5381,
"mean_token_accuracy": 0.15698456019163132,
"num_tokens": 14098192.0,
"step": 8125
},
{
"entropy": 5.657534599304199,
"epoch": 0.6325617584127602,
"grad_norm": 1.4765625,
"learning_rate": 0.0004965387238951536,
"loss": 5.36,
"mean_token_accuracy": 0.17074259370565414,
"num_tokens": 14106552.0,
"step": 8130
},
{
"entropy": 5.592873573303223,
"epoch": 0.6329507877844778,
"grad_norm": 2.859375,
"learning_rate": 0.0004965338794955418,
"loss": 5.4319,
"mean_token_accuracy": 0.16148714870214462,
"num_tokens": 14114783.0,
"step": 8135
},
{
"entropy": 5.535675430297852,
"epoch": 0.6333398171561953,
"grad_norm": 1.375,
"learning_rate": 0.0004965290317345061,
"loss": 5.4973,
"mean_token_accuracy": 0.15594300478696824,
"num_tokens": 14124182.0,
"step": 8140
},
{
"entropy": 5.678386974334717,
"epoch": 0.6337288465279128,
"grad_norm": 1.265625,
"learning_rate": 0.0004965241806121198,
"loss": 5.5245,
"mean_token_accuracy": 0.15432342439889907,
"num_tokens": 14133037.0,
"step": 8145
},
{
"entropy": 5.651481246948242,
"epoch": 0.6341178758996304,
"grad_norm": 1.46875,
"learning_rate": 0.0004965193261284567,
"loss": 5.4997,
"mean_token_accuracy": 0.16438781023025512,
"num_tokens": 14141677.0,
"step": 8150
},
{
"entropy": 5.569592905044556,
"epoch": 0.634506905271348,
"grad_norm": 1.453125,
"learning_rate": 0.0004965144682835904,
"loss": 5.4629,
"mean_token_accuracy": 0.17072515040636063,
"num_tokens": 14149980.0,
"step": 8155
},
{
"entropy": 5.563594961166382,
"epoch": 0.6348959346430656,
"grad_norm": 1.4296875,
"learning_rate": 0.0004965096070775946,
"loss": 5.5003,
"mean_token_accuracy": 0.16639333367347717,
"num_tokens": 14158298.0,
"step": 8160
},
{
"entropy": 5.627389335632325,
"epoch": 0.6352849640147831,
"grad_norm": 1.3359375,
"learning_rate": 0.0004965047425105431,
"loss": 5.4598,
"mean_token_accuracy": 0.16781965643167496,
"num_tokens": 14166321.0,
"step": 8165
},
{
"entropy": 5.5292116641998295,
"epoch": 0.6356739933865007,
"grad_norm": 1.3671875,
"learning_rate": 0.0004964998745825097,
"loss": 5.4246,
"mean_token_accuracy": 0.17231572419404984,
"num_tokens": 14174619.0,
"step": 8170
},
{
"entropy": 5.566997861862182,
"epoch": 0.6360630227582182,
"grad_norm": 1.4453125,
"learning_rate": 0.0004964950032935682,
"loss": 5.4403,
"mean_token_accuracy": 0.16089098602533342,
"num_tokens": 14183548.0,
"step": 8175
},
{
"entropy": 5.698858404159546,
"epoch": 0.6364520521299358,
"grad_norm": 1.546875,
"learning_rate": 0.0004964901286437927,
"loss": 5.6397,
"mean_token_accuracy": 0.1611291542649269,
"num_tokens": 14192306.0,
"step": 8180
},
{
"entropy": 5.5540143013000485,
"epoch": 0.6368410815016534,
"grad_norm": 1.46875,
"learning_rate": 0.0004964852506332568,
"loss": 5.3126,
"mean_token_accuracy": 0.1775052949786186,
"num_tokens": 14200280.0,
"step": 8185
},
{
"entropy": 5.570341730117798,
"epoch": 0.6372301108733709,
"grad_norm": 1.484375,
"learning_rate": 0.000496480369262035,
"loss": 5.5486,
"mean_token_accuracy": 0.16719427704811096,
"num_tokens": 14208484.0,
"step": 8190
},
{
"entropy": 5.606416893005371,
"epoch": 0.6376191402450885,
"grad_norm": 1.46875,
"learning_rate": 0.000496475484530201,
"loss": 5.5412,
"mean_token_accuracy": 0.16160168945789338,
"num_tokens": 14217368.0,
"step": 8195
},
{
"entropy": 5.693037462234497,
"epoch": 0.6380081696168061,
"grad_norm": 1.421875,
"learning_rate": 0.0004964705964378292,
"loss": 5.5571,
"mean_token_accuracy": 0.15757941454648972,
"num_tokens": 14226663.0,
"step": 8200
},
{
"entropy": 5.59523377418518,
"epoch": 0.6383971989885237,
"grad_norm": 1.4765625,
"learning_rate": 0.0004964657049849934,
"loss": 5.4228,
"mean_token_accuracy": 0.15858400166034697,
"num_tokens": 14235153.0,
"step": 8205
},
{
"entropy": 5.599985122680664,
"epoch": 0.6387862283602412,
"grad_norm": 1.546875,
"learning_rate": 0.0004964608101717681,
"loss": 5.5548,
"mean_token_accuracy": 0.15804918855428696,
"num_tokens": 14244245.0,
"step": 8210
},
{
"entropy": 5.63315052986145,
"epoch": 0.6391752577319587,
"grad_norm": 1.53125,
"learning_rate": 0.0004964559119982275,
"loss": 5.4708,
"mean_token_accuracy": 0.16142986565828324,
"num_tokens": 14252666.0,
"step": 8215
},
{
"entropy": 5.637704515457154,
"epoch": 0.6395642871036763,
"grad_norm": 1.3671875,
"learning_rate": 0.000496451010464446,
"loss": 5.548,
"mean_token_accuracy": 0.15910727083683013,
"num_tokens": 14262232.0,
"step": 8220
},
{
"entropy": 5.6048393726348875,
"epoch": 0.6399533164753939,
"grad_norm": 1.6875,
"learning_rate": 0.0004964461055704978,
"loss": 5.4765,
"mean_token_accuracy": 0.16455696076154708,
"num_tokens": 14270496.0,
"step": 8225
},
{
"entropy": 5.528129291534424,
"epoch": 0.6403423458471115,
"grad_norm": 1.5546875,
"learning_rate": 0.0004964411973164574,
"loss": 5.3994,
"mean_token_accuracy": 0.17100196778774263,
"num_tokens": 14278597.0,
"step": 8230
},
{
"entropy": 5.5415791988372805,
"epoch": 0.6407313752188291,
"grad_norm": 1.4453125,
"learning_rate": 0.0004964362857023993,
"loss": 5.4861,
"mean_token_accuracy": 0.16359479427337648,
"num_tokens": 14287444.0,
"step": 8235
},
{
"entropy": 5.563477563858032,
"epoch": 0.6411204045905465,
"grad_norm": 1.3984375,
"learning_rate": 0.000496431370728398,
"loss": 5.3847,
"mean_token_accuracy": 0.16639602482318877,
"num_tokens": 14296041.0,
"step": 8240
},
{
"entropy": 5.57078709602356,
"epoch": 0.6415094339622641,
"grad_norm": 1.34375,
"learning_rate": 0.0004964264523945281,
"loss": 5.4103,
"mean_token_accuracy": 0.1668490380048752,
"num_tokens": 14304700.0,
"step": 8245
},
{
"entropy": 5.613706636428833,
"epoch": 0.6418984633339817,
"grad_norm": 1.3671875,
"learning_rate": 0.0004964215307008643,
"loss": 5.5295,
"mean_token_accuracy": 0.16124773621559144,
"num_tokens": 14312873.0,
"step": 8250
},
{
"entropy": 5.674438238143921,
"epoch": 0.6422874927056993,
"grad_norm": 1.359375,
"learning_rate": 0.0004964166056474811,
"loss": 5.5317,
"mean_token_accuracy": 0.15775930359959603,
"num_tokens": 14321697.0,
"step": 8255
},
{
"entropy": 5.587344217300415,
"epoch": 0.6426765220774169,
"grad_norm": 1.4765625,
"learning_rate": 0.0004964116772344534,
"loss": 5.527,
"mean_token_accuracy": 0.1574459046125412,
"num_tokens": 14330074.0,
"step": 8260
},
{
"entropy": 5.696241664886474,
"epoch": 0.6430655514491345,
"grad_norm": 1.4765625,
"learning_rate": 0.0004964067454618559,
"loss": 5.5967,
"mean_token_accuracy": 0.15248970463871955,
"num_tokens": 14339485.0,
"step": 8265
},
{
"entropy": 5.717200422286988,
"epoch": 0.6434545808208519,
"grad_norm": 1.4296875,
"learning_rate": 0.0004964018103297634,
"loss": 5.5406,
"mean_token_accuracy": 0.15757163912057875,
"num_tokens": 14348182.0,
"step": 8270
},
{
"entropy": 5.728037929534912,
"epoch": 0.6438436101925695,
"grad_norm": 1.3046875,
"learning_rate": 0.0004963968718382509,
"loss": 5.6075,
"mean_token_accuracy": 0.155861759185791,
"num_tokens": 14357202.0,
"step": 8275
},
{
"entropy": 5.627114486694336,
"epoch": 0.6442326395642871,
"grad_norm": 1.40625,
"learning_rate": 0.0004963919299873931,
"loss": 5.5137,
"mean_token_accuracy": 0.15722453445196152,
"num_tokens": 14366001.0,
"step": 8280
},
{
"entropy": 5.695006942749023,
"epoch": 0.6446216689360047,
"grad_norm": 1.65625,
"learning_rate": 0.0004963869847772653,
"loss": 5.5558,
"mean_token_accuracy": 0.1576986774802208,
"num_tokens": 14374894.0,
"step": 8285
},
{
"entropy": 5.590862274169922,
"epoch": 0.6450106983077223,
"grad_norm": 1.7734375,
"learning_rate": 0.0004963820362079423,
"loss": 5.4925,
"mean_token_accuracy": 0.16458533555269242,
"num_tokens": 14383396.0,
"step": 8290
},
{
"entropy": 5.581786012649536,
"epoch": 0.6453997276794398,
"grad_norm": 1.546875,
"learning_rate": 0.0004963770842794994,
"loss": 5.4504,
"mean_token_accuracy": 0.1657285675406456,
"num_tokens": 14392514.0,
"step": 8295
},
{
"entropy": 5.629509305953979,
"epoch": 0.6457887570511573,
"grad_norm": 1.484375,
"learning_rate": 0.0004963721289920115,
"loss": 5.4895,
"mean_token_accuracy": 0.1577771633863449,
"num_tokens": 14401916.0,
"step": 8300
},
{
"entropy": 5.642601203918457,
"epoch": 0.6461777864228749,
"grad_norm": 1.671875,
"learning_rate": 0.0004963671703455538,
"loss": 5.5519,
"mean_token_accuracy": 0.15847557485103608,
"num_tokens": 14410476.0,
"step": 8305
},
{
"entropy": 5.65356411933899,
"epoch": 0.6465668157945925,
"grad_norm": 1.578125,
"learning_rate": 0.0004963622083402018,
"loss": 5.5281,
"mean_token_accuracy": 0.15930469781160356,
"num_tokens": 14418437.0,
"step": 8310
},
{
"entropy": 5.576390409469605,
"epoch": 0.6469558451663101,
"grad_norm": 1.515625,
"learning_rate": 0.0004963572429760305,
"loss": 5.4747,
"mean_token_accuracy": 0.16562929451465608,
"num_tokens": 14426978.0,
"step": 8315
},
{
"entropy": 5.662410306930542,
"epoch": 0.6473448745380276,
"grad_norm": 1.5234375,
"learning_rate": 0.0004963522742531155,
"loss": 5.5256,
"mean_token_accuracy": 0.15617482513189315,
"num_tokens": 14435043.0,
"step": 8320
},
{
"entropy": 5.5775915622711185,
"epoch": 0.6477339039097452,
"grad_norm": 1.515625,
"learning_rate": 0.000496347302171532,
"loss": 5.4221,
"mean_token_accuracy": 0.1707099497318268,
"num_tokens": 14443397.0,
"step": 8325
},
{
"entropy": 5.630001449584961,
"epoch": 0.6481229332814628,
"grad_norm": 1.5,
"learning_rate": 0.0004963423267313554,
"loss": 5.5111,
"mean_token_accuracy": 0.16315706968307495,
"num_tokens": 14452536.0,
"step": 8330
},
{
"entropy": 5.678388071060181,
"epoch": 0.6485119626531803,
"grad_norm": 2.28125,
"learning_rate": 0.0004963373479326614,
"loss": 5.5626,
"mean_token_accuracy": 0.152607698738575,
"num_tokens": 14461901.0,
"step": 8335
},
{
"entropy": 5.605925178527832,
"epoch": 0.6489009920248979,
"grad_norm": 1.6484375,
"learning_rate": 0.0004963323657755255,
"loss": 5.4965,
"mean_token_accuracy": 0.16920969188213347,
"num_tokens": 14470547.0,
"step": 8340
},
{
"entropy": 5.545820569992065,
"epoch": 0.6492900213966154,
"grad_norm": 1.5078125,
"learning_rate": 0.0004963273802600231,
"loss": 5.3666,
"mean_token_accuracy": 0.16293006539344787,
"num_tokens": 14478439.0,
"step": 8345
},
{
"entropy": 5.640049695968628,
"epoch": 0.649679050768333,
"grad_norm": 1.53125,
"learning_rate": 0.00049632239138623,
"loss": 5.4395,
"mean_token_accuracy": 0.1675925999879837,
"num_tokens": 14486621.0,
"step": 8350
},
{
"entropy": 5.583245944976807,
"epoch": 0.6500680801400506,
"grad_norm": 1.484375,
"learning_rate": 0.0004963173991542219,
"loss": 5.4247,
"mean_token_accuracy": 0.16209669709205626,
"num_tokens": 14495313.0,
"step": 8355
},
{
"entropy": 5.61328010559082,
"epoch": 0.6504571095117682,
"grad_norm": 1.5625,
"learning_rate": 0.0004963124035640746,
"loss": 5.4962,
"mean_token_accuracy": 0.16555400639772416,
"num_tokens": 14503647.0,
"step": 8360
},
{
"entropy": 5.612566041946411,
"epoch": 0.6508461388834857,
"grad_norm": 1.796875,
"learning_rate": 0.0004963074046158637,
"loss": 5.4302,
"mean_token_accuracy": 0.16509283930063248,
"num_tokens": 14512350.0,
"step": 8365
},
{
"entropy": 5.679213905334473,
"epoch": 0.6512351682552032,
"grad_norm": 1.5,
"learning_rate": 0.0004963024023096652,
"loss": 5.5734,
"mean_token_accuracy": 0.15210946649312973,
"num_tokens": 14521801.0,
"step": 8370
},
{
"entropy": 5.560119771957398,
"epoch": 0.6516241976269208,
"grad_norm": 1.59375,
"learning_rate": 0.0004962973966455551,
"loss": 5.4078,
"mean_token_accuracy": 0.16019269973039627,
"num_tokens": 14530045.0,
"step": 8375
},
{
"entropy": 5.55653567314148,
"epoch": 0.6520132269986384,
"grad_norm": 1.90625,
"learning_rate": 0.0004962923876236092,
"loss": 5.5038,
"mean_token_accuracy": 0.16818015426397323,
"num_tokens": 14538944.0,
"step": 8380
},
{
"entropy": 5.6275276184082035,
"epoch": 0.652402256370356,
"grad_norm": 1.5,
"learning_rate": 0.0004962873752439035,
"loss": 5.5462,
"mean_token_accuracy": 0.15992224514484404,
"num_tokens": 14548180.0,
"step": 8385
},
{
"entropy": 5.65324182510376,
"epoch": 0.6527912857420736,
"grad_norm": 1.6015625,
"learning_rate": 0.0004962823595065141,
"loss": 5.5374,
"mean_token_accuracy": 0.15571261197328568,
"num_tokens": 14556907.0,
"step": 8390
},
{
"entropy": 5.607539129257202,
"epoch": 0.653180315113791,
"grad_norm": 1.5078125,
"learning_rate": 0.0004962773404115172,
"loss": 5.5153,
"mean_token_accuracy": 0.15769898295402526,
"num_tokens": 14565317.0,
"step": 8395
},
{
"entropy": 5.683702993392944,
"epoch": 0.6535693444855086,
"grad_norm": 1.546875,
"learning_rate": 0.0004962723179589886,
"loss": 5.5783,
"mean_token_accuracy": 0.16089919656515123,
"num_tokens": 14574485.0,
"step": 8400
},
{
"entropy": 5.707424640655518,
"epoch": 0.6539583738572262,
"grad_norm": 1.640625,
"learning_rate": 0.000496267292149005,
"loss": 5.4845,
"mean_token_accuracy": 0.16174585819244386,
"num_tokens": 14584002.0,
"step": 8405
},
{
"entropy": 5.633489513397217,
"epoch": 0.6543474032289438,
"grad_norm": 1.5703125,
"learning_rate": 0.0004962622629816423,
"loss": 5.4311,
"mean_token_accuracy": 0.16070465072989465,
"num_tokens": 14592033.0,
"step": 8410
},
{
"entropy": 5.594094085693359,
"epoch": 0.6547364326006614,
"grad_norm": 1.6640625,
"learning_rate": 0.000496257230456977,
"loss": 5.5511,
"mean_token_accuracy": 0.15814293622970582,
"num_tokens": 14600084.0,
"step": 8415
},
{
"entropy": 5.609566974639892,
"epoch": 0.6551254619723789,
"grad_norm": 1.71875,
"learning_rate": 0.0004962521945750855,
"loss": 5.4655,
"mean_token_accuracy": 0.17209602296352386,
"num_tokens": 14608380.0,
"step": 8420
},
{
"entropy": 5.62038254737854,
"epoch": 0.6555144913440965,
"grad_norm": 1.5234375,
"learning_rate": 0.000496247155336044,
"loss": 5.5372,
"mean_token_accuracy": 0.16135943531990052,
"num_tokens": 14616864.0,
"step": 8425
},
{
"entropy": 5.643423366546631,
"epoch": 0.655903520715814,
"grad_norm": 1.828125,
"learning_rate": 0.0004962421127399291,
"loss": 5.5848,
"mean_token_accuracy": 0.15594003945589066,
"num_tokens": 14627054.0,
"step": 8430
},
{
"entropy": 5.681075525283814,
"epoch": 0.6562925500875316,
"grad_norm": 1.8125,
"learning_rate": 0.0004962370667868172,
"loss": 5.4622,
"mean_token_accuracy": 0.16358790993690492,
"num_tokens": 14634985.0,
"step": 8435
},
{
"entropy": 5.67454514503479,
"epoch": 0.6566815794592492,
"grad_norm": 1.9453125,
"learning_rate": 0.0004962320174767851,
"loss": 5.5275,
"mean_token_accuracy": 0.16311112195253372,
"num_tokens": 14644135.0,
"step": 8440
},
{
"entropy": 5.603046178817749,
"epoch": 0.6570706088309667,
"grad_norm": 1.6328125,
"learning_rate": 0.000496226964809909,
"loss": 5.475,
"mean_token_accuracy": 0.1671380341053009,
"num_tokens": 14652997.0,
"step": 8445
},
{
"entropy": 5.582294082641601,
"epoch": 0.6574596382026843,
"grad_norm": 1.515625,
"learning_rate": 0.000496221908786266,
"loss": 5.5694,
"mean_token_accuracy": 0.15573946237564087,
"num_tokens": 14661961.0,
"step": 8450
},
{
"entropy": 5.72095365524292,
"epoch": 0.6578486675744019,
"grad_norm": 1.4921875,
"learning_rate": 0.0004962168494059327,
"loss": 5.6857,
"mean_token_accuracy": 0.14969259947538377,
"num_tokens": 14671407.0,
"step": 8455
},
{
"entropy": 5.648622035980225,
"epoch": 0.6582376969461194,
"grad_norm": 3.46875,
"learning_rate": 0.0004962117866689857,
"loss": 5.5337,
"mean_token_accuracy": 0.16302797496318816,
"num_tokens": 14680015.0,
"step": 8460
},
{
"entropy": 5.558050012588501,
"epoch": 0.658626726317837,
"grad_norm": 1.65625,
"learning_rate": 0.000496206720575502,
"loss": 5.3794,
"mean_token_accuracy": 0.16196016669273378,
"num_tokens": 14688444.0,
"step": 8465
},
{
"entropy": 5.565975618362427,
"epoch": 0.6590157556895546,
"grad_norm": 1.6328125,
"learning_rate": 0.0004962016511255584,
"loss": 5.5212,
"mean_token_accuracy": 0.15961693376302719,
"num_tokens": 14697246.0,
"step": 8470
},
{
"entropy": 5.590129041671753,
"epoch": 0.6594047850612721,
"grad_norm": 1.9609375,
"learning_rate": 0.0004961965783192318,
"loss": 5.3933,
"mean_token_accuracy": 0.16685593873262405,
"num_tokens": 14706066.0,
"step": 8475
},
{
"entropy": 5.638283777236938,
"epoch": 0.6597938144329897,
"grad_norm": 1.3515625,
"learning_rate": 0.0004961915021565992,
"loss": 5.5079,
"mean_token_accuracy": 0.1568120762705803,
"num_tokens": 14714327.0,
"step": 8480
},
{
"entropy": 5.570411491394043,
"epoch": 0.6601828438047073,
"grad_norm": 1.5,
"learning_rate": 0.0004961864226377377,
"loss": 5.382,
"mean_token_accuracy": 0.16485304087400438,
"num_tokens": 14723053.0,
"step": 8485
},
{
"entropy": 5.544371938705444,
"epoch": 0.6605718731764249,
"grad_norm": 1.484375,
"learning_rate": 0.0004961813397627241,
"loss": 5.3785,
"mean_token_accuracy": 0.17082611471414566,
"num_tokens": 14731825.0,
"step": 8490
},
{
"entropy": 5.629836750030518,
"epoch": 0.6609609025481424,
"grad_norm": 1.609375,
"learning_rate": 0.0004961762535316358,
"loss": 5.5762,
"mean_token_accuracy": 0.156722342222929,
"num_tokens": 14740751.0,
"step": 8495
},
{
"entropy": 5.5798173427581785,
"epoch": 0.6613499319198599,
"grad_norm": 1.6953125,
"learning_rate": 0.0004961711639445499,
"loss": 5.4425,
"mean_token_accuracy": 0.16466536670923232,
"num_tokens": 14748994.0,
"step": 8500
},
{
"entropy": 5.582650995254516,
"epoch": 0.6617389612915775,
"grad_norm": 2.21875,
"learning_rate": 0.0004961660710015435,
"loss": 5.4294,
"mean_token_accuracy": 0.1654239133000374,
"num_tokens": 14757388.0,
"step": 8505
},
{
"entropy": 5.707845401763916,
"epoch": 0.6621279906632951,
"grad_norm": 1.5859375,
"learning_rate": 0.0004961609747026943,
"loss": 5.6107,
"mean_token_accuracy": 0.15208702832460402,
"num_tokens": 14766654.0,
"step": 8510
},
{
"entropy": 5.628918075561524,
"epoch": 0.6625170200350127,
"grad_norm": 1.4140625,
"learning_rate": 0.0004961558750480791,
"loss": 5.5065,
"mean_token_accuracy": 0.16062942296266555,
"num_tokens": 14775561.0,
"step": 8515
},
{
"entropy": 5.590156745910645,
"epoch": 0.6629060494067303,
"grad_norm": 1.59375,
"learning_rate": 0.0004961507720377756,
"loss": 5.5583,
"mean_token_accuracy": 0.15312558114528657,
"num_tokens": 14783596.0,
"step": 8520
},
{
"entropy": 5.673126602172852,
"epoch": 0.6632950787784477,
"grad_norm": 1.8984375,
"learning_rate": 0.0004961456656718611,
"loss": 5.5673,
"mean_token_accuracy": 0.1590380698442459,
"num_tokens": 14791932.0,
"step": 8525
},
{
"entropy": 5.685076332092285,
"epoch": 0.6636841081501653,
"grad_norm": 1.453125,
"learning_rate": 0.000496140555950413,
"loss": 5.6037,
"mean_token_accuracy": 0.15631160587072374,
"num_tokens": 14801261.0,
"step": 8530
},
{
"entropy": 5.622698163986206,
"epoch": 0.6640731375218829,
"grad_norm": 1.4921875,
"learning_rate": 0.0004961354428735091,
"loss": 5.4777,
"mean_token_accuracy": 0.1577353447675705,
"num_tokens": 14809788.0,
"step": 8535
},
{
"entropy": 5.542088317871094,
"epoch": 0.6644621668936005,
"grad_norm": 1.40625,
"learning_rate": 0.0004961303264412267,
"loss": 5.4653,
"mean_token_accuracy": 0.15726249366998674,
"num_tokens": 14818324.0,
"step": 8540
},
{
"entropy": 5.527442979812622,
"epoch": 0.6648511962653181,
"grad_norm": 1.8984375,
"learning_rate": 0.0004961252066536437,
"loss": 5.4235,
"mean_token_accuracy": 0.16347662657499312,
"num_tokens": 14827020.0,
"step": 8545
},
{
"entropy": 5.569663619995117,
"epoch": 0.6652402256370356,
"grad_norm": 1.6796875,
"learning_rate": 0.0004961200835108375,
"loss": 5.5391,
"mean_token_accuracy": 0.15555160492658615,
"num_tokens": 14835522.0,
"step": 8550
},
{
"entropy": 5.627895498275757,
"epoch": 0.6656292550087531,
"grad_norm": 1.9453125,
"learning_rate": 0.000496114957012886,
"loss": 5.493,
"mean_token_accuracy": 0.16153236776590346,
"num_tokens": 14844676.0,
"step": 8555
},
{
"entropy": 5.699510955810547,
"epoch": 0.6660182843804707,
"grad_norm": 1.6328125,
"learning_rate": 0.0004961098271598671,
"loss": 5.5413,
"mean_token_accuracy": 0.1609079509973526,
"num_tokens": 14853996.0,
"step": 8560
},
{
"entropy": 5.619197607040405,
"epoch": 0.6664073137521883,
"grad_norm": 1.671875,
"learning_rate": 0.0004961046939518585,
"loss": 5.5239,
"mean_token_accuracy": 0.15910777151584626,
"num_tokens": 14862981.0,
"step": 8565
},
{
"entropy": 5.658803033828735,
"epoch": 0.6667963431239059,
"grad_norm": 1.4921875,
"learning_rate": 0.0004960995573889379,
"loss": 5.5612,
"mean_token_accuracy": 0.1559648483991623,
"num_tokens": 14872378.0,
"step": 8570
},
{
"entropy": 5.589079093933106,
"epoch": 0.6671853724956234,
"grad_norm": 1.5078125,
"learning_rate": 0.0004960944174711836,
"loss": 5.4968,
"mean_token_accuracy": 0.15561144798994064,
"num_tokens": 14881662.0,
"step": 8575
},
{
"entropy": 5.621028661727905,
"epoch": 0.667574401867341,
"grad_norm": 1.9921875,
"learning_rate": 0.0004960892741986734,
"loss": 5.3817,
"mean_token_accuracy": 0.16623928993940354,
"num_tokens": 14890211.0,
"step": 8580
},
{
"entropy": 5.671985101699829,
"epoch": 0.6679634312390585,
"grad_norm": 1.4140625,
"learning_rate": 0.0004960841275714853,
"loss": 5.6131,
"mean_token_accuracy": 0.15232166945934295,
"num_tokens": 14899865.0,
"step": 8585
},
{
"entropy": 5.65290093421936,
"epoch": 0.6683524606107761,
"grad_norm": 1.5,
"learning_rate": 0.0004960789775896975,
"loss": 5.5242,
"mean_token_accuracy": 0.15458524227142334,
"num_tokens": 14908334.0,
"step": 8590
},
{
"entropy": 5.5436254978179935,
"epoch": 0.6687414899824937,
"grad_norm": 1.75,
"learning_rate": 0.0004960738242533881,
"loss": 5.4829,
"mean_token_accuracy": 0.16041945219039916,
"num_tokens": 14916477.0,
"step": 8595
},
{
"entropy": 5.554329109191895,
"epoch": 0.6691305193542112,
"grad_norm": 1.6796875,
"learning_rate": 0.0004960686675626353,
"loss": 5.4019,
"mean_token_accuracy": 0.16701223701238632,
"num_tokens": 14925223.0,
"step": 8600
},
{
"entropy": 5.687564706802368,
"epoch": 0.6695195487259288,
"grad_norm": 2.046875,
"learning_rate": 0.0004960635075175173,
"loss": 5.5495,
"mean_token_accuracy": 0.16739532649517058,
"num_tokens": 14934427.0,
"step": 8605
},
{
"entropy": 5.636126184463501,
"epoch": 0.6699085780976464,
"grad_norm": 1.515625,
"learning_rate": 0.0004960583441181124,
"loss": 5.4747,
"mean_token_accuracy": 0.16085093319416047,
"num_tokens": 14943175.0,
"step": 8610
},
{
"entropy": 5.587858009338379,
"epoch": 0.670297607469364,
"grad_norm": 1.5078125,
"learning_rate": 0.0004960531773644992,
"loss": 5.5039,
"mean_token_accuracy": 0.16135278195142747,
"num_tokens": 14951933.0,
"step": 8615
},
{
"entropy": 5.554653596878052,
"epoch": 0.6706866368410815,
"grad_norm": 3.734375,
"learning_rate": 0.0004960480072567557,
"loss": 5.491,
"mean_token_accuracy": 0.16200556010007858,
"num_tokens": 14960735.0,
"step": 8620
},
{
"entropy": 5.607791566848755,
"epoch": 0.671075666212799,
"grad_norm": 1.65625,
"learning_rate": 0.0004960428337949604,
"loss": 5.5836,
"mean_token_accuracy": 0.15506957173347474,
"num_tokens": 14970326.0,
"step": 8625
},
{
"entropy": 5.645353841781616,
"epoch": 0.6714646955845166,
"grad_norm": 1.6875,
"learning_rate": 0.0004960376569791921,
"loss": 5.4494,
"mean_token_accuracy": 0.1644549399614334,
"num_tokens": 14979822.0,
"step": 8630
},
{
"entropy": 5.733426570892334,
"epoch": 0.6718537249562342,
"grad_norm": 1.6640625,
"learning_rate": 0.0004960324768095291,
"loss": 5.5556,
"mean_token_accuracy": 0.15823704600334168,
"num_tokens": 14989350.0,
"step": 8635
},
{
"entropy": 5.657193660736084,
"epoch": 0.6722427543279518,
"grad_norm": 1.96875,
"learning_rate": 0.0004960272932860502,
"loss": 5.5569,
"mean_token_accuracy": 0.15302964970469474,
"num_tokens": 14998763.0,
"step": 8640
},
{
"entropy": 5.610429859161377,
"epoch": 0.6726317836996694,
"grad_norm": 1.8203125,
"learning_rate": 0.0004960221064088338,
"loss": 5.4387,
"mean_token_accuracy": 0.16410010904073716,
"num_tokens": 15006497.0,
"step": 8645
},
{
"entropy": 5.521879577636719,
"epoch": 0.6730208130713868,
"grad_norm": 2.140625,
"learning_rate": 0.0004960169161779588,
"loss": 5.4088,
"mean_token_accuracy": 0.16455425322055817,
"num_tokens": 15014397.0,
"step": 8650
},
{
"entropy": 5.5825025081634525,
"epoch": 0.6734098424431044,
"grad_norm": 1.4609375,
"learning_rate": 0.0004960117225935038,
"loss": 5.434,
"mean_token_accuracy": 0.1634689003229141,
"num_tokens": 15022726.0,
"step": 8655
},
{
"entropy": 5.689628839492798,
"epoch": 0.673798871814822,
"grad_norm": 1.40625,
"learning_rate": 0.0004960065256555477,
"loss": 5.6392,
"mean_token_accuracy": 0.15156934410333633,
"num_tokens": 15032317.0,
"step": 8660
},
{
"entropy": 5.632172536849976,
"epoch": 0.6741879011865396,
"grad_norm": 1.5390625,
"learning_rate": 0.0004960013253641695,
"loss": 5.5093,
"mean_token_accuracy": 0.16222840100526809,
"num_tokens": 15041091.0,
"step": 8665
},
{
"entropy": 5.696352577209472,
"epoch": 0.6745769305582572,
"grad_norm": 1.4921875,
"learning_rate": 0.0004959961217194477,
"loss": 5.6481,
"mean_token_accuracy": 0.15019654110074043,
"num_tokens": 15049797.0,
"step": 8670
},
{
"entropy": 5.562292814254761,
"epoch": 0.6749659599299748,
"grad_norm": 1.5078125,
"learning_rate": 0.0004959909147214615,
"loss": 5.3326,
"mean_token_accuracy": 0.17687424272298813,
"num_tokens": 15057736.0,
"step": 8675
},
{
"entropy": 5.586552238464355,
"epoch": 0.6753549893016922,
"grad_norm": 1.453125,
"learning_rate": 0.0004959857043702902,
"loss": 5.5146,
"mean_token_accuracy": 0.16430850476026534,
"num_tokens": 15066072.0,
"step": 8680
},
{
"entropy": 5.605910587310791,
"epoch": 0.6757440186734098,
"grad_norm": 1.3984375,
"learning_rate": 0.0004959804906660124,
"loss": 5.4266,
"mean_token_accuracy": 0.1668688103556633,
"num_tokens": 15073918.0,
"step": 8685
},
{
"entropy": 5.665558671951294,
"epoch": 0.6761330480451274,
"grad_norm": 1.4921875,
"learning_rate": 0.0004959752736087073,
"loss": 5.5073,
"mean_token_accuracy": 0.16089159101247788,
"num_tokens": 15082937.0,
"step": 8690
},
{
"entropy": 5.609713792800903,
"epoch": 0.676522077416845,
"grad_norm": 1.65625,
"learning_rate": 0.0004959700531984543,
"loss": 5.4734,
"mean_token_accuracy": 0.1653521776199341,
"num_tokens": 15091218.0,
"step": 8695
},
{
"entropy": 5.616780757904053,
"epoch": 0.6769111067885626,
"grad_norm": 1.5390625,
"learning_rate": 0.0004959648294353324,
"loss": 5.4071,
"mean_token_accuracy": 0.16858331859111786,
"num_tokens": 15099640.0,
"step": 8700
},
{
"entropy": 5.62099781036377,
"epoch": 0.6773001361602801,
"grad_norm": 1.375,
"learning_rate": 0.0004959596023194208,
"loss": 5.4814,
"mean_token_accuracy": 0.1604107677936554,
"num_tokens": 15108710.0,
"step": 8705
},
{
"entropy": 5.589712762832642,
"epoch": 0.6776891655319977,
"grad_norm": 1.328125,
"learning_rate": 0.000495954371850799,
"loss": 5.4417,
"mean_token_accuracy": 0.16216772943735122,
"num_tokens": 15117074.0,
"step": 8710
},
{
"entropy": 5.635924339294434,
"epoch": 0.6780781949037152,
"grad_norm": 1.453125,
"learning_rate": 0.0004959491380295463,
"loss": 5.4247,
"mean_token_accuracy": 0.1658407837152481,
"num_tokens": 15124714.0,
"step": 8715
},
{
"entropy": 5.545988988876343,
"epoch": 0.6784672242754328,
"grad_norm": 1.40625,
"learning_rate": 0.0004959439008557422,
"loss": 5.448,
"mean_token_accuracy": 0.16205866634845734,
"num_tokens": 15134177.0,
"step": 8720
},
{
"entropy": 5.591541147232055,
"epoch": 0.6788562536471504,
"grad_norm": 1.890625,
"learning_rate": 0.000495938660329466,
"loss": 5.4609,
"mean_token_accuracy": 0.16577730029821397,
"num_tokens": 15142659.0,
"step": 8725
},
{
"entropy": 5.62673921585083,
"epoch": 0.6792452830188679,
"grad_norm": 1.625,
"learning_rate": 0.0004959334164507973,
"loss": 5.6093,
"mean_token_accuracy": 0.15286325365304948,
"num_tokens": 15151423.0,
"step": 8730
},
{
"entropy": 5.624324989318848,
"epoch": 0.6796343123905855,
"grad_norm": 1.421875,
"learning_rate": 0.0004959281692198155,
"loss": 5.4109,
"mean_token_accuracy": 0.16259230971336364,
"num_tokens": 15159986.0,
"step": 8735
},
{
"entropy": 5.607788467407227,
"epoch": 0.6800233417623031,
"grad_norm": 1.421875,
"learning_rate": 0.0004959229186366007,
"loss": 5.4614,
"mean_token_accuracy": 0.16272899210453035,
"num_tokens": 15168183.0,
"step": 8740
},
{
"entropy": 5.509155893325806,
"epoch": 0.6804123711340206,
"grad_norm": 1.359375,
"learning_rate": 0.000495917664701232,
"loss": 5.3957,
"mean_token_accuracy": 0.1645374119281769,
"num_tokens": 15176476.0,
"step": 8745
},
{
"entropy": 5.751757669448852,
"epoch": 0.6808014005057382,
"grad_norm": 1.3671875,
"learning_rate": 0.0004959124074137894,
"loss": 5.6686,
"mean_token_accuracy": 0.15058834254741668,
"num_tokens": 15186873.0,
"step": 8750
},
{
"entropy": 5.654232454299927,
"epoch": 0.6811904298774557,
"grad_norm": 1.375,
"learning_rate": 0.0004959071467743526,
"loss": 5.4605,
"mean_token_accuracy": 0.1574256755411625,
"num_tokens": 15194965.0,
"step": 8755
},
{
"entropy": 5.511550617218018,
"epoch": 0.6815794592491733,
"grad_norm": 1.515625,
"learning_rate": 0.0004959018827830016,
"loss": 5.3325,
"mean_token_accuracy": 0.16597196459770203,
"num_tokens": 15204028.0,
"step": 8760
},
{
"entropy": 5.483450841903687,
"epoch": 0.6819684886208909,
"grad_norm": 1.5859375,
"learning_rate": 0.000495896615439816,
"loss": 5.3448,
"mean_token_accuracy": 0.17626579105854034,
"num_tokens": 15212011.0,
"step": 8765
},
{
"entropy": 5.540477132797241,
"epoch": 0.6823575179926085,
"grad_norm": 1.703125,
"learning_rate": 0.0004958913447448759,
"loss": 5.4426,
"mean_token_accuracy": 0.1627766102552414,
"num_tokens": 15220088.0,
"step": 8770
},
{
"entropy": 5.6379650115966795,
"epoch": 0.682746547364326,
"grad_norm": 1.5390625,
"learning_rate": 0.0004958860706982613,
"loss": 5.3944,
"mean_token_accuracy": 0.16646525710821153,
"num_tokens": 15228676.0,
"step": 8775
},
{
"entropy": 5.5093952178955075,
"epoch": 0.6831355767360435,
"grad_norm": 1.5390625,
"learning_rate": 0.000495880793300052,
"loss": 5.4235,
"mean_token_accuracy": 0.168090882897377,
"num_tokens": 15237392.0,
"step": 8780
},
{
"entropy": 5.555091094970703,
"epoch": 0.6835246061077611,
"grad_norm": 1.6875,
"learning_rate": 0.0004958755125503284,
"loss": 5.563,
"mean_token_accuracy": 0.1629531815648079,
"num_tokens": 15246738.0,
"step": 8785
},
{
"entropy": 5.70695013999939,
"epoch": 0.6839136354794787,
"grad_norm": 1.421875,
"learning_rate": 0.0004958702284491703,
"loss": 5.5492,
"mean_token_accuracy": 0.1600048288702965,
"num_tokens": 15255154.0,
"step": 8790
},
{
"entropy": 5.617724895477295,
"epoch": 0.6843026648511963,
"grad_norm": 1.53125,
"learning_rate": 0.0004958649409966581,
"loss": 5.4238,
"mean_token_accuracy": 0.1639481157064438,
"num_tokens": 15264178.0,
"step": 8795
},
{
"entropy": 5.6335736274719235,
"epoch": 0.6846916942229139,
"grad_norm": 1.625,
"learning_rate": 0.000495859650192872,
"loss": 5.5289,
"mean_token_accuracy": 0.1640607163310051,
"num_tokens": 15272591.0,
"step": 8800
},
{
"entropy": 5.627588701248169,
"epoch": 0.6850807235946313,
"grad_norm": 1.3828125,
"learning_rate": 0.0004958543560378922,
"loss": 5.4459,
"mean_token_accuracy": 0.16031597703695297,
"num_tokens": 15281280.0,
"step": 8805
},
{
"entropy": 5.577515888214111,
"epoch": 0.6854697529663489,
"grad_norm": 1.5703125,
"learning_rate": 0.0004958490585317991,
"loss": 5.3933,
"mean_token_accuracy": 0.169253808259964,
"num_tokens": 15290639.0,
"step": 8810
},
{
"entropy": 5.539572143554688,
"epoch": 0.6858587823380665,
"grad_norm": 1.546875,
"learning_rate": 0.000495843757674673,
"loss": 5.3904,
"mean_token_accuracy": 0.15841607749462128,
"num_tokens": 15298877.0,
"step": 8815
},
{
"entropy": 5.627851152420044,
"epoch": 0.6862478117097841,
"grad_norm": 1.40625,
"learning_rate": 0.0004958384534665945,
"loss": 5.5185,
"mean_token_accuracy": 0.15833221822977067,
"num_tokens": 15307710.0,
"step": 8820
},
{
"entropy": 5.643237590789795,
"epoch": 0.6866368410815017,
"grad_norm": 1.5859375,
"learning_rate": 0.0004958331459076438,
"loss": 5.5354,
"mean_token_accuracy": 0.16657649874687194,
"num_tokens": 15315749.0,
"step": 8825
},
{
"entropy": 5.666751956939697,
"epoch": 0.6870258704532192,
"grad_norm": 1.5625,
"learning_rate": 0.0004958278349979018,
"loss": 5.4937,
"mean_token_accuracy": 0.15831152200698853,
"num_tokens": 15324161.0,
"step": 8830
},
{
"entropy": 5.525657415390015,
"epoch": 0.6874148998249368,
"grad_norm": 1.65625,
"learning_rate": 0.0004958225207374488,
"loss": 5.3865,
"mean_token_accuracy": 0.16515809148550034,
"num_tokens": 15333414.0,
"step": 8835
},
{
"entropy": 5.619464302062989,
"epoch": 0.6878039291966543,
"grad_norm": 1.484375,
"learning_rate": 0.0004958172031263655,
"loss": 5.4721,
"mean_token_accuracy": 0.1637744888663292,
"num_tokens": 15342369.0,
"step": 8840
},
{
"entropy": 5.668105411529541,
"epoch": 0.6881929585683719,
"grad_norm": 1.5234375,
"learning_rate": 0.0004958118821647326,
"loss": 5.5448,
"mean_token_accuracy": 0.154964642226696,
"num_tokens": 15351109.0,
"step": 8845
},
{
"entropy": 5.5900016784667965,
"epoch": 0.6885819879400895,
"grad_norm": 1.609375,
"learning_rate": 0.0004958065578526308,
"loss": 5.4351,
"mean_token_accuracy": 0.16864016205072402,
"num_tokens": 15359837.0,
"step": 8850
},
{
"entropy": 5.547193765640259,
"epoch": 0.688971017311807,
"grad_norm": 1.9921875,
"learning_rate": 0.000495801230190141,
"loss": 5.4663,
"mean_token_accuracy": 0.1670744426548481,
"num_tokens": 15368776.0,
"step": 8855
},
{
"entropy": 5.571122741699218,
"epoch": 0.6893600466835246,
"grad_norm": 1.5703125,
"learning_rate": 0.0004957958991773441,
"loss": 5.4472,
"mean_token_accuracy": 0.16593606919050216,
"num_tokens": 15378127.0,
"step": 8860
},
{
"entropy": 5.618199396133423,
"epoch": 0.6897490760552422,
"grad_norm": 1.53125,
"learning_rate": 0.0004957905648143206,
"loss": 5.4145,
"mean_token_accuracy": 0.1654287099838257,
"num_tokens": 15386580.0,
"step": 8865
},
{
"entropy": 5.626070880889893,
"epoch": 0.6901381054269597,
"grad_norm": 1.4765625,
"learning_rate": 0.0004957852271011519,
"loss": 5.5675,
"mean_token_accuracy": 0.1599985420703888,
"num_tokens": 15395973.0,
"step": 8870
},
{
"entropy": 5.6116798400878904,
"epoch": 0.6905271347986773,
"grad_norm": 1.5,
"learning_rate": 0.0004957798860379187,
"loss": 5.4506,
"mean_token_accuracy": 0.1576478213071823,
"num_tokens": 15404738.0,
"step": 8875
},
{
"entropy": 5.624153327941895,
"epoch": 0.6909161641703949,
"grad_norm": 1.484375,
"learning_rate": 0.0004957745416247022,
"loss": 5.5947,
"mean_token_accuracy": 0.1643390730023384,
"num_tokens": 15413530.0,
"step": 8880
},
{
"entropy": 5.61382155418396,
"epoch": 0.6913051935421124,
"grad_norm": 1.375,
"learning_rate": 0.0004957691938615833,
"loss": 5.5804,
"mean_token_accuracy": 0.15752760097384452,
"num_tokens": 15423268.0,
"step": 8885
},
{
"entropy": 5.573393630981445,
"epoch": 0.69169422291383,
"grad_norm": 1.53125,
"learning_rate": 0.0004957638427486434,
"loss": 5.3878,
"mean_token_accuracy": 0.1690160557627678,
"num_tokens": 15431690.0,
"step": 8890
},
{
"entropy": 5.602618646621704,
"epoch": 0.6920832522855476,
"grad_norm": 1.421875,
"learning_rate": 0.0004957584882859636,
"loss": 5.5484,
"mean_token_accuracy": 0.15811869204044343,
"num_tokens": 15440110.0,
"step": 8895
},
{
"entropy": 5.714527559280396,
"epoch": 0.6924722816572652,
"grad_norm": 1.6875,
"learning_rate": 0.0004957531304736251,
"loss": 5.5609,
"mean_token_accuracy": 0.15900706499814987,
"num_tokens": 15449022.0,
"step": 8900
},
{
"entropy": 5.74059796333313,
"epoch": 0.6928613110289827,
"grad_norm": 1.3828125,
"learning_rate": 0.0004957477693117091,
"loss": 5.6081,
"mean_token_accuracy": 0.1521860897541046,
"num_tokens": 15457826.0,
"step": 8905
},
{
"entropy": 5.633170747756958,
"epoch": 0.6932503404007002,
"grad_norm": 1.4453125,
"learning_rate": 0.0004957424048002971,
"loss": 5.4866,
"mean_token_accuracy": 0.16168588250875474,
"num_tokens": 15466631.0,
"step": 8910
},
{
"entropy": 5.528276968002319,
"epoch": 0.6936393697724178,
"grad_norm": 1.296875,
"learning_rate": 0.0004957370369394706,
"loss": 5.3796,
"mean_token_accuracy": 0.16295055896043778,
"num_tokens": 15474666.0,
"step": 8915
},
{
"entropy": 5.515926742553711,
"epoch": 0.6940283991441354,
"grad_norm": 1.625,
"learning_rate": 0.0004957316657293107,
"loss": 5.3883,
"mean_token_accuracy": 0.16150261461734772,
"num_tokens": 15483576.0,
"step": 8920
},
{
"entropy": 5.744505786895752,
"epoch": 0.694417428515853,
"grad_norm": 1.5390625,
"learning_rate": 0.0004957262911698992,
"loss": 5.5505,
"mean_token_accuracy": 0.15730631202459336,
"num_tokens": 15491313.0,
"step": 8925
},
{
"entropy": 5.61945481300354,
"epoch": 0.6948064578875706,
"grad_norm": 1.3671875,
"learning_rate": 0.0004957209132613175,
"loss": 5.5039,
"mean_token_accuracy": 0.16225907653570176,
"num_tokens": 15500638.0,
"step": 8930
},
{
"entropy": 5.5722874164581295,
"epoch": 0.695195487259288,
"grad_norm": 1.4375,
"learning_rate": 0.0004957155320036473,
"loss": 5.4996,
"mean_token_accuracy": 0.16195370107889176,
"num_tokens": 15509170.0,
"step": 8935
},
{
"entropy": 5.737703943252564,
"epoch": 0.6955845166310056,
"grad_norm": 1.3828125,
"learning_rate": 0.0004957101473969702,
"loss": 5.5868,
"mean_token_accuracy": 0.1541694298386574,
"num_tokens": 15518108.0,
"step": 8940
},
{
"entropy": 5.608725214004517,
"epoch": 0.6959735460027232,
"grad_norm": 1.5,
"learning_rate": 0.000495704759441368,
"loss": 5.4399,
"mean_token_accuracy": 0.16826948374509812,
"num_tokens": 15526558.0,
"step": 8945
},
{
"entropy": 5.567080926895142,
"epoch": 0.6963625753744408,
"grad_norm": 1.4609375,
"learning_rate": 0.0004956993681369221,
"loss": 5.4086,
"mean_token_accuracy": 0.16608458757400513,
"num_tokens": 15535685.0,
"step": 8950
},
{
"entropy": 5.598057079315185,
"epoch": 0.6967516047461584,
"grad_norm": 1.5390625,
"learning_rate": 0.0004956939734837148,
"loss": 5.4573,
"mean_token_accuracy": 0.16753346621990203,
"num_tokens": 15544072.0,
"step": 8955
},
{
"entropy": 5.616986417770386,
"epoch": 0.6971406341178759,
"grad_norm": 1.546875,
"learning_rate": 0.0004956885754818277,
"loss": 5.5587,
"mean_token_accuracy": 0.1520525261759758,
"num_tokens": 15553303.0,
"step": 8960
},
{
"entropy": 5.633440065383911,
"epoch": 0.6975296634895934,
"grad_norm": 1.359375,
"learning_rate": 0.0004956831741313427,
"loss": 5.4995,
"mean_token_accuracy": 0.15973107665777206,
"num_tokens": 15561942.0,
"step": 8965
},
{
"entropy": 5.623577308654785,
"epoch": 0.697918692861311,
"grad_norm": 1.5,
"learning_rate": 0.0004956777694323418,
"loss": 5.3534,
"mean_token_accuracy": 0.172841776907444,
"num_tokens": 15569988.0,
"step": 8970
},
{
"entropy": 5.570699214935303,
"epoch": 0.6983077222330286,
"grad_norm": 1.703125,
"learning_rate": 0.000495672361384907,
"loss": 5.5355,
"mean_token_accuracy": 0.16330803483724593,
"num_tokens": 15579816.0,
"step": 8975
},
{
"entropy": 5.595700073242187,
"epoch": 0.6986967516047462,
"grad_norm": 2.03125,
"learning_rate": 0.0004956669499891202,
"loss": 5.4168,
"mean_token_accuracy": 0.1699569582939148,
"num_tokens": 15587843.0,
"step": 8980
},
{
"entropy": 5.651184034347534,
"epoch": 0.6990857809764637,
"grad_norm": 1.5859375,
"learning_rate": 0.0004956615352450639,
"loss": 5.4735,
"mean_token_accuracy": 0.16514448672533036,
"num_tokens": 15596798.0,
"step": 8985
},
{
"entropy": 5.618335151672364,
"epoch": 0.6994748103481813,
"grad_norm": 1.4609375,
"learning_rate": 0.0004956561171528198,
"loss": 5.5535,
"mean_token_accuracy": 0.16553656309843062,
"num_tokens": 15605360.0,
"step": 8990
},
{
"entropy": 5.550785064697266,
"epoch": 0.6998638397198989,
"grad_norm": 1.546875,
"learning_rate": 0.0004956506957124704,
"loss": 5.4225,
"mean_token_accuracy": 0.170002281665802,
"num_tokens": 15613557.0,
"step": 8995
},
{
"entropy": 5.597484922409057,
"epoch": 0.7002528690916164,
"grad_norm": 1.3984375,
"learning_rate": 0.000495645270924098,
"loss": 5.4867,
"mean_token_accuracy": 0.16351661831140518,
"num_tokens": 15622144.0,
"step": 9000
},
{
"epoch": 0.7002528690916164,
"eval_entropy": 5.496048769615841,
"eval_loss": 5.52293062210083,
"eval_mean_token_accuracy": 0.16872901462226447,
"eval_num_tokens": 15622144.0,
"eval_runtime": 28.4802,
"eval_samples_per_second": 1459.192,
"eval_steps_per_second": 182.408,
"step": 9000
}
],
"logging_steps": 5,
"max_steps": 128520,
"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": 2.1228899770368e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink