{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.277337559429477,
"global_step": 700,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 3.846153846153847e-06,
"loss": 2.7825,
"theoretical_loss": 3.3518567762894107,
"tokens_seen": 2969305088
},
{
"epoch": 0.0,
"learning_rate": 3.846153846153846e-05,
"loss": 2.4663,
"theoretical_loss": 3.3517550749686795,
"tokens_seen": 2970484736
},
{
"epoch": 0.01,
"learning_rate": 7.692307692307693e-05,
"loss": 1.9063,
"theoretical_loss": 3.351642134115535,
"tokens_seen": 2971795456
},
{
"epoch": 0.01,
"learning_rate": 9.983987189751803e-05,
"loss": 1.4439,
"theoretical_loss": 3.351529257004948,
"tokens_seen": 2973106176
},
{
"epoch": 0.02,
"learning_rate": 9.943955164131305e-05,
"loss": 1.2731,
"theoretical_loss": 3.3514164435728655,
"tokens_seen": 2974416896
},
{
"epoch": 0.02,
"learning_rate": 9.90392313851081e-05,
"loss": 1.2079,
"theoretical_loss": 3.3513036937553267,
"tokens_seen": 2975727616
},
{
"epoch": 0.02,
"learning_rate": 9.863891112890312e-05,
"loss": 1.2085,
"theoretical_loss": 3.3511910074884628,
"tokens_seen": 2977038336
},
{
"epoch": 0.03,
"learning_rate": 9.823859087269817e-05,
"loss": 1.2031,
"theoretical_loss": 3.3510783847084977,
"tokens_seen": 2978349056
},
{
"epoch": 0.03,
"learning_rate": 9.78382706164932e-05,
"loss": 1.2176,
"theoretical_loss": 3.350965825351748,
"tokens_seen": 2979659776
},
{
"epoch": 0.04,
"learning_rate": 9.743795036028824e-05,
"loss": 1.1891,
"theoretical_loss": 3.3508533293546208,
"tokens_seen": 2980970496
},
{
"epoch": 0.04,
"learning_rate": 9.703763010408326e-05,
"loss": 1.2086,
"theoretical_loss": 3.3507408966536154,
"tokens_seen": 2982281216
},
{
"epoch": 0.04,
"learning_rate": 9.663730984787832e-05,
"loss": 1.18,
"theoretical_loss": 3.3506285271853233,
"tokens_seen": 2983591936
},
{
"epoch": 0.05,
"learning_rate": 9.623698959167334e-05,
"loss": 1.1709,
"theoretical_loss": 3.350516220886427,
"tokens_seen": 2984902656
},
{
"epoch": 0.05,
"objective/train/docs_used": 1000073,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.024436354637146,
"objective/train/theoretical_loss": 3.3504600914057403,
"objective/train/tokens_used": 36842976,
"theoretical_loss": 3.3504600914057403,
"tokens_seen": 2985558016
},
{
"epoch": 0.05,
"learning_rate": 9.583666933546838e-05,
"loss": 1.1659,
"theoretical_loss": 3.3504039776936994,
"tokens_seen": 2986213376
},
{
"epoch": 0.06,
"learning_rate": 9.543634907926342e-05,
"loss": 1.1826,
"theoretical_loss": 3.350291797544005,
"tokens_seen": 2987524096
},
{
"epoch": 0.06,
"learning_rate": 9.503602882305846e-05,
"loss": 1.1854,
"theoretical_loss": 3.3501796803742994,
"tokens_seen": 2988834816
},
{
"epoch": 0.06,
"learning_rate": 9.463570856685348e-05,
"loss": 1.1901,
"theoretical_loss": 3.3500676261216285,
"tokens_seen": 2990145536
},
{
"epoch": 0.07,
"learning_rate": 9.423538831064852e-05,
"loss": 1.1854,
"theoretical_loss": 3.3499556347231287,
"tokens_seen": 2991456256
},
{
"epoch": 0.07,
"learning_rate": 9.383506805444356e-05,
"loss": 1.2042,
"theoretical_loss": 3.349843706116027,
"tokens_seen": 2992766976
},
{
"epoch": 0.08,
"learning_rate": 9.34347477982386e-05,
"loss": 1.1804,
"theoretical_loss": 3.3497318402376397,
"tokens_seen": 2994077696
},
{
"epoch": 0.08,
"learning_rate": 9.303442754203363e-05,
"loss": 1.1869,
"theoretical_loss": 3.349620037025374,
"tokens_seen": 2995388416
},
{
"epoch": 0.08,
"learning_rate": 9.263410728582867e-05,
"loss": 1.1783,
"theoretical_loss": 3.349508296416727,
"tokens_seen": 2996699136
},
{
"epoch": 0.09,
"learning_rate": 9.22337870296237e-05,
"loss": 1.1659,
"theoretical_loss": 3.3493966183492847,
"tokens_seen": 2998009856
},
{
"epoch": 0.09,
"learning_rate": 9.183346677341874e-05,
"loss": 1.1763,
"theoretical_loss": 3.349285002760723,
"tokens_seen": 2999320576
},
{
"epoch": 0.1,
"learning_rate": 9.143314651721377e-05,
"loss": 1.151,
"theoretical_loss": 3.3491734495888066,
"tokens_seen": 3000631296
},
{
"epoch": 0.1,
"objective/train/docs_used": 1005438,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.0497773885726929,
"objective/train/theoretical_loss": 3.3490619587713906,
"objective/train/tokens_used": 53226976,
"theoretical_loss": 3.3490619587713906,
"tokens_seen": 3001942016
},
{
"epoch": 0.1,
"learning_rate": 9.103282626100881e-05,
"loss": 1.1792,
"theoretical_loss": 3.3490619587713906,
"tokens_seen": 3001942016
},
{
"epoch": 0.1,
"learning_rate": 9.063250600480385e-05,
"loss": 1.1536,
"theoretical_loss": 3.348950530246418,
"tokens_seen": 3003252736
},
{
"epoch": 0.11,
"learning_rate": 9.023218574859889e-05,
"loss": 1.1929,
"theoretical_loss": 3.348839163951921,
"tokens_seen": 3004563456
},
{
"epoch": 0.11,
"learning_rate": 8.983186549239391e-05,
"loss": 1.1682,
"theoretical_loss": 3.3487278598260195,
"tokens_seen": 3005874176
},
{
"epoch": 0.11,
"learning_rate": 8.943154523618895e-05,
"loss": 1.148,
"theoretical_loss": 3.348616617806923,
"tokens_seen": 3007184896
},
{
"epoch": 0.12,
"learning_rate": 8.903122497998399e-05,
"loss": 1.1768,
"theoretical_loss": 3.34850543783293,
"tokens_seen": 3008495616
},
{
"epoch": 0.12,
"learning_rate": 8.863090472377903e-05,
"loss": 1.1853,
"theoretical_loss": 3.3483943198424253,
"tokens_seen": 3009806336
},
{
"epoch": 0.13,
"learning_rate": 8.823058446757406e-05,
"loss": 1.1791,
"theoretical_loss": 3.348283263773882,
"tokens_seen": 3011117056
},
{
"epoch": 0.13,
"learning_rate": 8.78302642113691e-05,
"loss": 1.18,
"theoretical_loss": 3.3481722695658633,
"tokens_seen": 3012427776
},
{
"epoch": 0.13,
"learning_rate": 8.742994395516414e-05,
"loss": 1.1761,
"theoretical_loss": 3.3480613371570165,
"tokens_seen": 3013738496
},
{
"epoch": 0.14,
"learning_rate": 8.702962369895917e-05,
"loss": 1.1834,
"theoretical_loss": 3.347950466486079,
"tokens_seen": 3015049216
},
{
"epoch": 0.14,
"learning_rate": 8.66293034427542e-05,
"loss": 1.1986,
"theoretical_loss": 3.3478396574918747,
"tokens_seen": 3016359936
},
{
"epoch": 0.15,
"learning_rate": 8.622898318654924e-05,
"loss": 1.1895,
"theoretical_loss": 3.347728910113315,
"tokens_seen": 3017670656
},
{
"epoch": 0.15,
"objective/train/docs_used": 1011213,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.2239086627960205,
"objective/train/theoretical_loss": 3.3476735595108353,
"objective/train/tokens_used": 69610976,
"theoretical_loss": 3.3476735595108353,
"tokens_seen": 3018326016
},
{
"epoch": 0.15,
"learning_rate": 8.582866293034428e-05,
"loss": 1.1745,
"theoretical_loss": 3.3476182242893975,
"tokens_seen": 3018981376
},
{
"epoch": 0.15,
"learning_rate": 8.542834267413932e-05,
"loss": 1.1915,
"theoretical_loss": 3.3475075999592074,
"tokens_seen": 3020292096
},
{
"epoch": 0.16,
"learning_rate": 8.502802241793434e-05,
"loss": 1.1925,
"theoretical_loss": 3.347397037061916,
"tokens_seen": 3021602816
},
{
"epoch": 0.16,
"learning_rate": 8.462770216172938e-05,
"loss": 1.1872,
"theoretical_loss": 3.3472865355367825,
"tokens_seen": 3022913536
},
{
"epoch": 0.17,
"learning_rate": 8.422738190552442e-05,
"loss": 1.1753,
"theoretical_loss": 3.3471760953231504,
"tokens_seen": 3024224256
},
{
"epoch": 0.17,
"learning_rate": 8.382706164931946e-05,
"loss": 1.2088,
"theoretical_loss": 3.3470657163604507,
"tokens_seen": 3025534976
},
{
"epoch": 0.17,
"learning_rate": 8.34267413931145e-05,
"loss": 1.1959,
"theoretical_loss": 3.3469553985882,
"tokens_seen": 3026845696
},
{
"epoch": 0.18,
"learning_rate": 8.302642113690953e-05,
"loss": 1.2173,
"theoretical_loss": 3.3468451419460017,
"tokens_seen": 3028156416
},
{
"epoch": 0.18,
"learning_rate": 8.262610088070457e-05,
"loss": 1.1876,
"theoretical_loss": 3.3467349463735436,
"tokens_seen": 3029467136
},
{
"epoch": 0.19,
"learning_rate": 8.22257806244996e-05,
"loss": 1.1839,
"theoretical_loss": 3.3466248118105995,
"tokens_seen": 3030777856
},
{
"epoch": 0.19,
"learning_rate": 8.182546036829464e-05,
"loss": 1.198,
"theoretical_loss": 3.3465147381970293,
"tokens_seen": 3032088576
},
{
"epoch": 0.19,
"learning_rate": 8.142514011208967e-05,
"loss": 1.1891,
"theoretical_loss": 3.346404725472777,
"tokens_seen": 3033399296
},
{
"epoch": 0.2,
"objective/train/docs_used": 1016482,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 1.2322360277175903,
"objective/train/theoretical_loss": 3.3462947735778727,
"objective/train/tokens_used": 85994976,
"theoretical_loss": 3.3462947735778727,
"tokens_seen": 3034710016
},
{
"epoch": 0.2,
"learning_rate": 8.102481985588471e-05,
"loss": 1.1985,
"theoretical_loss": 3.3462947735778727,
"tokens_seen": 3034710016
},
{
"epoch": 0.2,
"learning_rate": 8.062449959967975e-05,
"loss": 1.1688,
"theoretical_loss": 3.346184882452431,
"tokens_seen": 3036020736
},
{
"epoch": 0.21,
"learning_rate": 8.022417934347479e-05,
"loss": 1.1874,
"theoretical_loss": 3.3460750520366513,
"tokens_seen": 3037331456
},
{
"epoch": 0.21,
"learning_rate": 7.982385908726981e-05,
"loss": 1.1835,
"theoretical_loss": 3.3459652822708175,
"tokens_seen": 3038642176
},
{
"epoch": 0.21,
"learning_rate": 7.942353883106487e-05,
"loss": 1.1842,
"theoretical_loss": 3.345855573095298,
"tokens_seen": 3039952896
},
{
"epoch": 0.22,
"learning_rate": 7.902321857485989e-05,
"loss": 1.1882,
"theoretical_loss": 3.3457459244505463,
"tokens_seen": 3041263616
},
{
"epoch": 0.22,
"learning_rate": 7.862289831865493e-05,
"loss": 1.2044,
"theoretical_loss": 3.3456363362770984,
"tokens_seen": 3042574336
},
{
"epoch": 0.23,
"learning_rate": 7.822257806244996e-05,
"loss": 1.168,
"theoretical_loss": 3.345526808515576,
"tokens_seen": 3043885056
},
{
"epoch": 0.23,
"learning_rate": 7.782225780624501e-05,
"loss": 1.1908,
"theoretical_loss": 3.3454173411066828,
"tokens_seen": 3045195776
},
{
"epoch": 0.23,
"learning_rate": 7.742193755004003e-05,
"loss": 1.1942,
"theoretical_loss": 3.345307933991209,
"tokens_seen": 3046506496
},
{
"epoch": 0.24,
"learning_rate": 7.702161729383507e-05,
"loss": 1.1965,
"theoretical_loss": 3.3451985871100254,
"tokens_seen": 3047817216
},
{
"epoch": 0.24,
"learning_rate": 7.662129703763011e-05,
"loss": 1.1727,
"theoretical_loss": 3.345089300404087,
"tokens_seen": 3049127936
},
{
"epoch": 0.25,
"learning_rate": 7.622097678142515e-05,
"loss": 1.1993,
"theoretical_loss": 3.344980073814434,
"tokens_seen": 3050438656
},
{
"epoch": 0.25,
"objective/train/docs_used": 1022217,
"objective/train/instantaneous_batch_size": 16,
"objective/train/instantaneous_microbatch_size": 16384,
"objective/train/original_loss": 0.9348351955413818,
"objective/train/theoretical_loss": 3.3449254830448116,
"objective/train/tokens_used": 102378976,
"theoretical_loss": 3.3449254830448116,
"tokens_seen": 3051094016
},
{
"epoch": 0.25,
"learning_rate": 7.582065652522018e-05,
"loss": 1.1904,
"theoretical_loss": 3.3448709072821874,
"tokens_seen": 3051749376
},
{
"epoch": 0.25,
"learning_rate": 7.542033626901522e-05,
"loss": 1.1939,
"theoretical_loss": 3.344761800748551,
"tokens_seen": 3053060096
},
{
"epoch": 0.26,
"learning_rate": 7.502001601281026e-05,
"loss": 1.1901,
"theoretical_loss": 3.3446527541548132,
"tokens_seen": 3054370816
},
{
"epoch": 0.26,
"learning_rate": 7.46196957566053e-05,
"loss": 1.2002,
"theoretical_loss": 3.3445437674423437,
"tokens_seen": 3055681536
},
{
"epoch": 0.27,
"learning_rate": 7.421937550040032e-05,
"loss": 1.2222,
"theoretical_loss": 3.3444348405525948,
"tokens_seen": 3056992256
},
{
"epoch": 0.27,
"learning_rate": 7.381905524419536e-05,
"loss": 1.2025,
"theoretical_loss": 3.3443259734271016,
"tokens_seen": 3058302976
},
{
"epoch": 0.27,
"learning_rate": 7.34187349879904e-05,
"loss": 1.1916,
"theoretical_loss": 3.344217166007481,
"tokens_seen": 3059613696
},
{
"epoch": 0.28,
"learning_rate": 7.301841473178544e-05,
"loss": 1.1965,
"theoretical_loss": 3.344108418235432,
"tokens_seen": 3060924416
}
],
"max_steps": 2524,
"num_train_epochs": 9223372036854775807,
"total_flos": 4.68235321344e+16,
"trial_name": null,
"trial_params": null
}