| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 78735, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0009936495840477551, | |
| "loss": 1.2096, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0009872991680955102, | |
| "loss": 1.2958, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0009809487521432654, | |
| "loss": 1.3278, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0009745983361910206, | |
| "loss": 1.3214, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0009682479202387756, | |
| "loss": 1.3299, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0009618975042865308, | |
| "loss": 1.3244, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0009555470883342859, | |
| "loss": 1.3279, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.000949196672382041, | |
| "loss": 1.3269, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0009428462564297961, | |
| "loss": 1.3179, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0009364958404775514, | |
| "loss": 1.3057, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.58219165847995, | |
| "eval_loss": 2.046381950378418, | |
| "eval_runtime": 1873.4069, | |
| "eval_samples_per_second": 89.652, | |
| "eval_steps_per_second": 0.701, | |
| "step": 5249 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0009301454245253064, | |
| "loss": 1.1726, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.0009237950085730616, | |
| "loss": 1.0732, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0009174445926208167, | |
| "loss": 1.0922, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.0009110941766685719, | |
| "loss": 1.1075, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.0009047437607163269, | |
| "loss": 1.112, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.0008983933447640821, | |
| "loss": 1.1417, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.0008920429288118372, | |
| "loss": 1.1343, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.0008856925128595922, | |
| "loss": 1.1319, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.0008793420969073475, | |
| "loss": 1.1326, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.0008729916809551026, | |
| "loss": 1.1387, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6084486916138251, | |
| "eval_loss": 1.9456982612609863, | |
| "eval_runtime": 4861.9841, | |
| "eval_samples_per_second": 34.545, | |
| "eval_steps_per_second": 0.27, | |
| "step": 10498 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0008666412650028577, | |
| "loss": 1.124, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0008602908490506128, | |
| "loss": 0.9, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.000853940433098368, | |
| "loss": 0.9401, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 0.000847590017146123, | |
| "loss": 0.967, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.0008412396011938782, | |
| "loss": 0.9734, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0008348891852416333, | |
| "loss": 0.9657, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 0.0008285387692893885, | |
| "loss": 0.9883, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.0008221883533371436, | |
| "loss": 0.9781, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.0008158379373848988, | |
| "loss": 0.9822, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 0.0008094875214326538, | |
| "loss": 0.992, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 0.000803137105480409, | |
| "loss": 0.988, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6124319014021613, | |
| "eval_loss": 1.965035319328308, | |
| "eval_runtime": 4587.2857, | |
| "eval_samples_per_second": 36.613, | |
| "eval_steps_per_second": 0.286, | |
| "step": 15747 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.0007967866895281641, | |
| "loss": 0.8822, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 0.0007904362735759192, | |
| "loss": 0.8149, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 0.0007840858576236743, | |
| "loss": 0.8325, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.0007777354416714295, | |
| "loss": 0.8412, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 0.0007713850257191847, | |
| "loss": 0.8421, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 0.0007650346097669397, | |
| "loss": 0.8487, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 0.0007586841938146949, | |
| "loss": 0.855, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.00075233377786245, | |
| "loss": 0.8623, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 0.0007459833619102051, | |
| "loss": 0.8668, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.0007396329459579602, | |
| "loss": 0.8653, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.634402071983567, | |
| "eval_loss": 1.9380866289138794, | |
| "eval_runtime": 1885.7719, | |
| "eval_samples_per_second": 89.064, | |
| "eval_steps_per_second": 0.696, | |
| "step": 20996 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0007332825300057155, | |
| "loss": 0.864, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 0.0007269321140534705, | |
| "loss": 0.7073, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.0007205816981012257, | |
| "loss": 0.7257, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 0.0007142312821489808, | |
| "loss": 0.7415, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 0.0007078808661967359, | |
| "loss": 0.7397, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.000701530450244491, | |
| "loss": 0.7635, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.0006951800342922462, | |
| "loss": 0.7736, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.0006888296183400012, | |
| "loss": 0.7694, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 0.0006824792023877565, | |
| "loss": 0.759, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 0.0006761287864355116, | |
| "loss": 0.7728, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.0006697783704832666, | |
| "loss": 0.7662, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6335327915215385, | |
| "eval_loss": 1.9391114711761475, | |
| "eval_runtime": 1717.5004, | |
| "eval_samples_per_second": 97.79, | |
| "eval_steps_per_second": 0.764, | |
| "step": 26245 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 0.0006634279545310218, | |
| "loss": 0.7034, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 0.0006570775385787769, | |
| "loss": 0.6446, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 0.000650727122626532, | |
| "loss": 0.6511, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.0006443767066742871, | |
| "loss": 0.6562, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.0006380262907220423, | |
| "loss": 0.6673, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 0.0006316758747697975, | |
| "loss": 0.6719, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 0.0006253254588175526, | |
| "loss": 0.6818, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 0.0006189750428653077, | |
| "loss": 0.6745, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.0006126246269130629, | |
| "loss": 0.6778, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.0006062742109608179, | |
| "loss": 0.6882, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6443749813938258, | |
| "eval_loss": 1.9590805768966675, | |
| "eval_runtime": 1673.9405, | |
| "eval_samples_per_second": 100.335, | |
| "eval_steps_per_second": 0.784, | |
| "step": 31494 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0005999237950085731, | |
| "loss": 0.6787, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "learning_rate": 0.0005935733790563282, | |
| "loss": 0.5613, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.0005872229631040833, | |
| "loss": 0.5867, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.0005808725471518384, | |
| "loss": 0.5839, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 0.0005745221311995937, | |
| "loss": 0.588, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 0.0005681717152473487, | |
| "loss": 0.6022, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 0.0005618212992951039, | |
| "loss": 0.596, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.000555470883342859, | |
| "loss": 0.6072, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 0.000549120467390614, | |
| "loss": 0.615, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 0.0005427700514383692, | |
| "loss": 0.6103, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 0.0005364196354861243, | |
| "loss": 0.601, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6510374802774552, | |
| "eval_loss": 1.9506298303604126, | |
| "eval_runtime": 4758.5372, | |
| "eval_samples_per_second": 35.296, | |
| "eval_steps_per_second": 0.276, | |
| "step": 36743 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 0.0005300692195338794, | |
| "loss": 0.5518, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 0.0005237188035816346, | |
| "loss": 0.5144, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 0.0005173683876293898, | |
| "loss": 0.5214, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 0.0005110179716771448, | |
| "loss": 0.522, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 0.0005046675557249, | |
| "loss": 0.5331, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 0.0004983171397726551, | |
| "loss": 0.5357, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 0.0004919667238204102, | |
| "loss": 0.53, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 0.00048561630786816534, | |
| "loss": 0.5397, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 0.0004792658919159205, | |
| "loss": 0.5319, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 0.0004729154759636756, | |
| "loss": 0.5363, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6616772349736536, | |
| "eval_loss": 1.9555561542510986, | |
| "eval_runtime": 4754.1095, | |
| "eval_samples_per_second": 35.328, | |
| "eval_steps_per_second": 0.276, | |
| "step": 41992 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.0004665650600114307, | |
| "loss": 0.3811, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 0.0004602146440591859, | |
| "loss": 0.4593, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 8.19, | |
| "learning_rate": 0.000453864228106941, | |
| "loss": 0.4769, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 0.00044751381215469617, | |
| "loss": 0.4676, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 0.0004411633962024513, | |
| "loss": 0.4816, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 0.0004348129802502064, | |
| "loss": 0.4765, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 0.00042846256429796155, | |
| "loss": 0.4835, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 0.00042211214834571666, | |
| "loss": 0.4767, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 0.0004157617323934718, | |
| "loss": 0.4877, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 0.00040941131644122694, | |
| "loss": 0.4706, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 0.00040306090048898205, | |
| "loss": 0.4871, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6740853204727457, | |
| "eval_loss": 1.9037331342697144, | |
| "eval_runtime": 5063.3946, | |
| "eval_samples_per_second": 33.17, | |
| "eval_steps_per_second": 0.259, | |
| "step": 47241 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 0.00039671048453673716, | |
| "loss": 0.4473, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "learning_rate": 0.0003903600685844923, | |
| "loss": 0.4062, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 0.0003840096526322474, | |
| "loss": 0.4208, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "learning_rate": 0.00037765923668000255, | |
| "loss": 0.4233, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 0.00037130882072775766, | |
| "loss": 0.4211, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 0.0003649584047755128, | |
| "loss": 0.4273, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 0.00035860798882326794, | |
| "loss": 0.4259, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 0.00035225757287102305, | |
| "loss": 0.4157, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "learning_rate": 0.00034590715691877816, | |
| "loss": 0.4294, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "learning_rate": 0.00033955674096653333, | |
| "loss": 0.4338, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6806287398410289, | |
| "eval_loss": 1.9794013500213623, | |
| "eval_runtime": 4548.1376, | |
| "eval_samples_per_second": 36.928, | |
| "eval_steps_per_second": 0.289, | |
| "step": 52490 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.00033320632501428844, | |
| "loss": 0.4278, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 10.1, | |
| "learning_rate": 0.00032685590906204355, | |
| "loss": 0.3655, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "learning_rate": 0.0003205054931097987, | |
| "loss": 0.3792, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 10.29, | |
| "learning_rate": 0.00031415507715755383, | |
| "loss": 0.3765, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 0.000307804661205309, | |
| "loss": 0.3777, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 0.0003014542452530641, | |
| "loss": 0.3766, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 10.57, | |
| "learning_rate": 0.0002951038293008192, | |
| "loss": 0.3847, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "learning_rate": 0.0002887534133485744, | |
| "loss": 0.3802, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 0.00028240299739632944, | |
| "loss": 0.3836, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "learning_rate": 0.00027605258144408455, | |
| "loss": 0.3696, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 10.95, | |
| "learning_rate": 0.0002697021654918397, | |
| "loss": 0.3738, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6849275103450329, | |
| "eval_loss": 2.0053181648254395, | |
| "eval_runtime": 3187.3586, | |
| "eval_samples_per_second": 52.694, | |
| "eval_steps_per_second": 0.412, | |
| "step": 57739 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 0.0002633517495395948, | |
| "loss": 0.3603, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "learning_rate": 0.00025700133358734994, | |
| "loss": 0.3371, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 0.0002506509176351051, | |
| "loss": 0.3334, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 11.34, | |
| "learning_rate": 0.0002443005016828602, | |
| "loss": 0.3342, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 0.00023795008573061535, | |
| "loss": 0.3392, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 11.53, | |
| "learning_rate": 0.0002315996697783705, | |
| "loss": 0.3323, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 0.00022524925382612563, | |
| "loss": 0.3444, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 0.00021889883787388074, | |
| "loss": 0.3395, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 11.81, | |
| "learning_rate": 0.00021254842192163588, | |
| "loss": 0.3321, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 11.91, | |
| "learning_rate": 0.00020619800596939102, | |
| "loss": 0.3338, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6955315411866274, | |
| "eval_loss": 2.0140492916107178, | |
| "eval_runtime": 1705.2241, | |
| "eval_samples_per_second": 98.494, | |
| "eval_steps_per_second": 0.77, | |
| "step": 62988 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.0001998475900171461, | |
| "loss": 0.3326, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.00019349717406490124, | |
| "loss": 0.3104, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 12.19, | |
| "learning_rate": 0.00018714675811265638, | |
| "loss": 0.2943, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "learning_rate": 0.00018079634216041152, | |
| "loss": 0.3063, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "learning_rate": 0.00017444592620816663, | |
| "loss": 0.302, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 0.00016809551025592177, | |
| "loss": 0.295, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 12.57, | |
| "learning_rate": 0.0001617450943036769, | |
| "loss": 0.2945, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 0.00015539467835143204, | |
| "loss": 0.3072, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 0.00014904426239918715, | |
| "loss": 0.2913, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 0.00014269384644694227, | |
| "loss": 0.3055, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 12.95, | |
| "learning_rate": 0.0001363434304946974, | |
| "loss": 0.2989, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6974427674079366, | |
| "eval_loss": 2.083035707473755, | |
| "eval_runtime": 1683.3073, | |
| "eval_samples_per_second": 99.777, | |
| "eval_steps_per_second": 0.78, | |
| "step": 68237 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 0.00012999301454245252, | |
| "loss": 0.2808, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 13.15, | |
| "learning_rate": 0.00012364259859020765, | |
| "loss": 0.2789, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 0.00011729218263796279, | |
| "loss": 0.2743, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "learning_rate": 0.00011094176668571792, | |
| "loss": 0.2767, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 13.43, | |
| "learning_rate": 0.00010459135073347304, | |
| "loss": 0.2648, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "learning_rate": 9.824093478122817e-05, | |
| "loss": 0.2724, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 13.62, | |
| "learning_rate": 9.18905188289833e-05, | |
| "loss": 0.2753, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 8.554010287673843e-05, | |
| "loss": 0.2743, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 13.81, | |
| "learning_rate": 7.918968692449357e-05, | |
| "loss": 0.2658, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 13.91, | |
| "learning_rate": 7.283927097224868e-05, | |
| "loss": 0.267, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7040516805096603, | |
| "eval_loss": 2.0823681354522705, | |
| "eval_runtime": 2701.2349, | |
| "eval_samples_per_second": 62.177, | |
| "eval_steps_per_second": 0.486, | |
| "step": 73486 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 6.64888550200038e-05, | |
| "loss": 0.2592, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 6.013843906775894e-05, | |
| "loss": 0.2473, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 5.378802311551407e-05, | |
| "loss": 0.2471, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 4.74376071632692e-05, | |
| "loss": 0.2592, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 14.38, | |
| "learning_rate": 4.1087191211024324e-05, | |
| "loss": 0.2493, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 3.473677525877945e-05, | |
| "loss": 0.2484, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "learning_rate": 2.838635930653458e-05, | |
| "loss": 0.2468, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 14.67, | |
| "learning_rate": 2.203594335428971e-05, | |
| "loss": 0.2432, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 1.5685527402044834e-05, | |
| "loss": 0.2466, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 14.86, | |
| "learning_rate": 9.335111449799962e-06, | |
| "loss": 0.2472, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "learning_rate": 2.98469549755509e-06, | |
| "loss": 0.236, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.708475484504778, | |
| "eval_loss": 2.0700299739837646, | |
| "eval_runtime": 1790.1892, | |
| "eval_samples_per_second": 93.82, | |
| "eval_steps_per_second": 0.733, | |
| "step": 78735 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 78735, | |
| "total_flos": 6.872831806674565e+20, | |
| "train_loss": 0.0, | |
| "train_runtime": 39.8476, | |
| "train_samples_per_second": 252894.596, | |
| "train_steps_per_second": 1975.901 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 78735, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "total_flos": 6.872831806674565e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |