| { | |
| "best_metric": 0.8533899487562311, | |
| "best_model_checkpoint": "./results/checkpoint-1200", | |
| "epoch": 69.0, | |
| "eval_steps": 500, | |
| "global_step": 1656, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.4191670417785645, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 1.1717, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.5559615812547036, | |
| "eval_loss": 0.852466344833374, | |
| "eval_runtime": 0.2518, | |
| "eval_samples_per_second": 865.826, | |
| "eval_steps_per_second": 15.887, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.028530597686768, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.714, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.7746990836466595, | |
| "eval_loss": 0.6411612033843994, | |
| "eval_runtime": 0.2498, | |
| "eval_samples_per_second": 872.757, | |
| "eval_steps_per_second": 16.014, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 6.396097660064697, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.4376, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.7904679173662312, | |
| "eval_loss": 0.578528881072998, | |
| "eval_runtime": 0.2466, | |
| "eval_samples_per_second": 883.847, | |
| "eval_steps_per_second": 16.217, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 13.205650329589844, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.262, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.8032839961396288, | |
| "eval_loss": 0.6008332967758179, | |
| "eval_runtime": 0.2457, | |
| "eval_samples_per_second": 887.351, | |
| "eval_steps_per_second": 16.282, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 9.777926445007324, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.1384, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.7854101778464291, | |
| "eval_loss": 0.6125118732452393, | |
| "eval_runtime": 0.2485, | |
| "eval_samples_per_second": 877.32, | |
| "eval_steps_per_second": 16.098, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.165589332580566, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.0982, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_f1": 0.7672888039973402, | |
| "eval_loss": 0.7653169631958008, | |
| "eval_runtime": 0.2569, | |
| "eval_samples_per_second": 848.674, | |
| "eval_steps_per_second": 15.572, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.7239115238189697, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.0657, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_f1": 0.8147995774570269, | |
| "eval_loss": 0.7002198100090027, | |
| "eval_runtime": 0.2474, | |
| "eval_samples_per_second": 881.237, | |
| "eval_steps_per_second": 16.169, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.3135942220687866, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.0395, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_f1": 0.8087986139563679, | |
| "eval_loss": 0.7771649956703186, | |
| "eval_runtime": 0.244, | |
| "eval_samples_per_second": 893.39, | |
| "eval_steps_per_second": 16.392, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.4529080092906952, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.0214, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_f1": 0.8392522702480626, | |
| "eval_loss": 0.7208238244056702, | |
| "eval_runtime": 0.2501, | |
| "eval_samples_per_second": 871.738, | |
| "eval_steps_per_second": 15.995, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 8.327765464782715, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0237, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_f1": 0.8424386724386725, | |
| "eval_loss": 0.7019856572151184, | |
| "eval_runtime": 0.2469, | |
| "eval_samples_per_second": 883.057, | |
| "eval_steps_per_second": 16.203, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.6359758377075195, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.0159, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_f1": 0.8074501800720288, | |
| "eval_loss": 0.9096614122390747, | |
| "eval_runtime": 0.2471, | |
| "eval_samples_per_second": 882.066, | |
| "eval_steps_per_second": 16.185, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 18.37872886657715, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.0322, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_f1": 0.7848399290035054, | |
| "eval_loss": 1.0800021886825562, | |
| "eval_runtime": 0.2488, | |
| "eval_samples_per_second": 876.232, | |
| "eval_steps_per_second": 16.078, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 13.307744026184082, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.029, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_f1": 0.798355010551357, | |
| "eval_loss": 1.0241613388061523, | |
| "eval_runtime": 0.2493, | |
| "eval_samples_per_second": 874.372, | |
| "eval_steps_per_second": 16.044, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.41816067695617676, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.0121, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_f1": 0.8120306519736268, | |
| "eval_loss": 1.022652268409729, | |
| "eval_runtime": 0.2454, | |
| "eval_samples_per_second": 888.194, | |
| "eval_steps_per_second": 16.297, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 10.153714179992676, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.0082, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_f1": 0.7906564533144493, | |
| "eval_loss": 1.096178650856018, | |
| "eval_runtime": 0.2461, | |
| "eval_samples_per_second": 885.781, | |
| "eval_steps_per_second": 16.253, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.026447944343090057, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.0035, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_f1": 0.7900391731356858, | |
| "eval_loss": 1.2207499742507935, | |
| "eval_runtime": 0.2461, | |
| "eval_samples_per_second": 885.87, | |
| "eval_steps_per_second": 16.254, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.008760841563344002, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.0064, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_f1": 0.7941827325696338, | |
| "eval_loss": 1.120672345161438, | |
| "eval_runtime": 0.2456, | |
| "eval_samples_per_second": 887.604, | |
| "eval_steps_per_second": 16.286, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.2175062894821167, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.0089, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_f1": 0.8121032920020557, | |
| "eval_loss": 1.1158227920532227, | |
| "eval_runtime": 0.2466, | |
| "eval_samples_per_second": 884.124, | |
| "eval_steps_per_second": 16.222, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.026284487918019295, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.0011, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_f1": 0.8106351612230018, | |
| "eval_loss": 1.1229100227355957, | |
| "eval_runtime": 0.2466, | |
| "eval_samples_per_second": 883.852, | |
| "eval_steps_per_second": 16.217, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.1518473625183105, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0036, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_f1": 0.8215225082145855, | |
| "eval_loss": 1.0700623989105225, | |
| "eval_runtime": 0.2447, | |
| "eval_samples_per_second": 890.866, | |
| "eval_steps_per_second": 16.346, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 0.030479425564408302, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.0038, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_f1": 0.79625307224709, | |
| "eval_loss": 1.1593633890151978, | |
| "eval_runtime": 0.2451, | |
| "eval_samples_per_second": 889.358, | |
| "eval_steps_per_second": 16.319, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 0.8992727994918823, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.013, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_f1": 0.8198287286484007, | |
| "eval_loss": 1.127288579940796, | |
| "eval_runtime": 0.2483, | |
| "eval_samples_per_second": 878.117, | |
| "eval_steps_per_second": 16.112, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 0.008566158823668957, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.0051, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_f1": 0.8245210270785887, | |
| "eval_loss": 1.0870376825332642, | |
| "eval_runtime": 0.2503, | |
| "eval_samples_per_second": 870.992, | |
| "eval_steps_per_second": 15.982, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 0.0786171406507492, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.0037, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_f1": 0.8210009216488835, | |
| "eval_loss": 1.1669812202453613, | |
| "eval_runtime": 0.2462, | |
| "eval_samples_per_second": 885.624, | |
| "eval_steps_per_second": 16.25, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 0.02009885385632515, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0035, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_f1": 0.823341564899699, | |
| "eval_loss": 1.1029468774795532, | |
| "eval_runtime": 0.246, | |
| "eval_samples_per_second": 886.238, | |
| "eval_steps_per_second": 16.261, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 0.10086794197559357, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.0051, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_f1": 0.8138320306415473, | |
| "eval_loss": 1.1733108758926392, | |
| "eval_runtime": 0.2405, | |
| "eval_samples_per_second": 906.567, | |
| "eval_steps_per_second": 16.634, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 24.029403686523438, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.0032, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_f1": 0.8293087494328629, | |
| "eval_loss": 1.1072059869766235, | |
| "eval_runtime": 0.244, | |
| "eval_samples_per_second": 893.364, | |
| "eval_steps_per_second": 16.392, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 19.329952239990234, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.024, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_f1": 0.7828629987222324, | |
| "eval_loss": 1.3776278495788574, | |
| "eval_runtime": 0.2463, | |
| "eval_samples_per_second": 884.934, | |
| "eval_steps_per_second": 16.237, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": 0.0337708480656147, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.0097, | |
| "step": 696 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_f1": 0.8245008772063647, | |
| "eval_loss": 1.161023736000061, | |
| "eval_runtime": 0.246, | |
| "eval_samples_per_second": 886.218, | |
| "eval_steps_per_second": 16.261, | |
| "step": 696 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.05386161431670189, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.0093, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_f1": 0.8331553578084018, | |
| "eval_loss": 1.0987571477890015, | |
| "eval_runtime": 0.2499, | |
| "eval_samples_per_second": 872.378, | |
| "eval_steps_per_second": 16.007, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "grad_norm": 0.011382571421563625, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.0048, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_f1": 0.8253722493039259, | |
| "eval_loss": 1.1277176141738892, | |
| "eval_runtime": 0.2469, | |
| "eval_samples_per_second": 882.906, | |
| "eval_steps_per_second": 16.2, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.0046822689473629, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.0041, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_f1": 0.8187101207450439, | |
| "eval_loss": 1.1826363801956177, | |
| "eval_runtime": 0.2504, | |
| "eval_samples_per_second": 870.593, | |
| "eval_steps_per_second": 15.974, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "grad_norm": 0.003356009954586625, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.0033, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_f1": 0.8126497250026661, | |
| "eval_loss": 1.1944386959075928, | |
| "eval_runtime": 0.2453, | |
| "eval_samples_per_second": 888.767, | |
| "eval_steps_per_second": 16.308, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "grad_norm": 0.005327207036316395, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.0087, | |
| "step": 816 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_f1": 0.8491252629565882, | |
| "eval_loss": 1.03484046459198, | |
| "eval_runtime": 0.2487, | |
| "eval_samples_per_second": 876.726, | |
| "eval_steps_per_second": 16.087, | |
| "step": 816 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 0.003193259472027421, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.0056, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_f1": 0.831069428937251, | |
| "eval_loss": 1.0799349546432495, | |
| "eval_runtime": 0.2433, | |
| "eval_samples_per_second": 895.888, | |
| "eval_steps_per_second": 16.438, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 0.0025981140788644552, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.0056, | |
| "step": 864 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_f1": 0.8005370004504828, | |
| "eval_loss": 1.2397780418395996, | |
| "eval_runtime": 0.2558, | |
| "eval_samples_per_second": 852.208, | |
| "eval_steps_per_second": 15.637, | |
| "step": 864 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "grad_norm": 0.004587420262396336, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.0043, | |
| "step": 888 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_f1": 0.7960422975955538, | |
| "eval_loss": 1.3035740852355957, | |
| "eval_runtime": 0.2431, | |
| "eval_samples_per_second": 896.654, | |
| "eval_steps_per_second": 16.452, | |
| "step": 888 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "grad_norm": 0.0024126123171299696, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.004, | |
| "step": 912 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_f1": 0.8348215283066768, | |
| "eval_loss": 1.1289474964141846, | |
| "eval_runtime": 0.2451, | |
| "eval_samples_per_second": 889.465, | |
| "eval_steps_per_second": 16.32, | |
| "step": 912 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "grad_norm": 0.002631419338285923, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.0002, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_f1": 0.8470719620834265, | |
| "eval_loss": 1.0962085723876953, | |
| "eval_runtime": 0.2555, | |
| "eval_samples_per_second": 853.196, | |
| "eval_steps_per_second": 15.655, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.0036367273423820734, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0002, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_f1": 0.8475274660470973, | |
| "eval_loss": 1.1172648668289185, | |
| "eval_runtime": 0.2665, | |
| "eval_samples_per_second": 818.091, | |
| "eval_steps_per_second": 15.011, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "grad_norm": 0.002203166950494051, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.0002, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_f1": 0.8463311286594244, | |
| "eval_loss": 1.1128673553466797, | |
| "eval_runtime": 0.2476, | |
| "eval_samples_per_second": 880.33, | |
| "eval_steps_per_second": 16.153, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "grad_norm": 0.0019745519384741783, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.0002, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_f1": 0.8524668435013261, | |
| "eval_loss": 1.1187357902526855, | |
| "eval_runtime": 0.2433, | |
| "eval_samples_per_second": 896.105, | |
| "eval_steps_per_second": 16.442, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "grad_norm": 0.0018844620790332556, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.0002, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_f1": 0.8524668435013261, | |
| "eval_loss": 1.1086227893829346, | |
| "eval_runtime": 0.2535, | |
| "eval_samples_per_second": 860.088, | |
| "eval_steps_per_second": 15.781, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "grad_norm": 0.0016616833163425326, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.0002, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_f1": 0.8524668435013261, | |
| "eval_loss": 1.1084070205688477, | |
| "eval_runtime": 0.2467, | |
| "eval_samples_per_second": 883.486, | |
| "eval_steps_per_second": 16.211, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 0.0017370691057294607, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.0002, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_f1": 0.8524668435013261, | |
| "eval_loss": 1.1120808124542236, | |
| "eval_runtime": 0.2544, | |
| "eval_samples_per_second": 856.909, | |
| "eval_steps_per_second": 15.723, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "grad_norm": 0.01709812693297863, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.0018, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_f1": 0.8230951784833656, | |
| "eval_loss": 1.150782585144043, | |
| "eval_runtime": 0.2507, | |
| "eval_samples_per_second": 869.395, | |
| "eval_steps_per_second": 15.952, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "grad_norm": 0.0014819600619375706, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.0033, | |
| "step": 1128 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_f1": 0.8482902683102467, | |
| "eval_loss": 1.1178399324417114, | |
| "eval_runtime": 0.251, | |
| "eval_samples_per_second": 868.684, | |
| "eval_steps_per_second": 15.939, | |
| "step": 1128 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 0.001657757442444563, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.0012, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_f1": 0.8485012469416702, | |
| "eval_loss": 1.1701571941375732, | |
| "eval_runtime": 0.2457, | |
| "eval_samples_per_second": 887.403, | |
| "eval_steps_per_second": 16.283, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "grad_norm": 0.001472759060561657, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.0001, | |
| "step": 1176 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_f1": 0.8471931986968837, | |
| "eval_loss": 1.1576160192489624, | |
| "eval_runtime": 0.2492, | |
| "eval_samples_per_second": 874.869, | |
| "eval_steps_per_second": 16.053, | |
| "step": 1176 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.0018886495381593704, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0001, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_f1": 0.8533899487562311, | |
| "eval_loss": 1.1580414772033691, | |
| "eval_runtime": 0.2453, | |
| "eval_samples_per_second": 888.766, | |
| "eval_steps_per_second": 16.308, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "grad_norm": 0.0017091715708374977, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.0001, | |
| "step": 1224 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_f1": 0.8533899487562311, | |
| "eval_loss": 1.157599687576294, | |
| "eval_runtime": 0.2534, | |
| "eval_samples_per_second": 860.384, | |
| "eval_steps_per_second": 15.787, | |
| "step": 1224 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "grad_norm": 0.0013613449409604073, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.0001, | |
| "step": 1248 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_f1": 0.8533899487562311, | |
| "eval_loss": 1.155985713005066, | |
| "eval_runtime": 0.2572, | |
| "eval_samples_per_second": 847.53, | |
| "eval_steps_per_second": 15.551, | |
| "step": 1248 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "grad_norm": 0.001274469424970448, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.0001, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_f1": 0.8533899487562311, | |
| "eval_loss": 1.1571167707443237, | |
| "eval_runtime": 0.2412, | |
| "eval_samples_per_second": 903.632, | |
| "eval_steps_per_second": 16.58, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "grad_norm": 0.0013310050126165152, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.0001, | |
| "step": 1296 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_f1": 0.8533899487562311, | |
| "eval_loss": 1.1600090265274048, | |
| "eval_runtime": 0.2473, | |
| "eval_samples_per_second": 881.514, | |
| "eval_steps_per_second": 16.175, | |
| "step": 1296 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 0.0015337098157033324, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.0001, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_f1": 0.8426159574795251, | |
| "eval_loss": 1.1735517978668213, | |
| "eval_runtime": 0.2491, | |
| "eval_samples_per_second": 875.11, | |
| "eval_steps_per_second": 16.057, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "grad_norm": 0.008407847955822945, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.0001, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_f1": 0.8426159574795251, | |
| "eval_loss": 1.1661510467529297, | |
| "eval_runtime": 0.2486, | |
| "eval_samples_per_second": 876.938, | |
| "eval_steps_per_second": 16.091, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "grad_norm": 0.0011946760350838304, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.0001, | |
| "step": 1368 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_f1": 0.8475855910205873, | |
| "eval_loss": 1.1671065092086792, | |
| "eval_runtime": 0.2535, | |
| "eval_samples_per_second": 859.876, | |
| "eval_steps_per_second": 15.778, | |
| "step": 1368 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "grad_norm": 0.0011800089851021767, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.0001, | |
| "step": 1392 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_f1": 0.8475855910205873, | |
| "eval_loss": 1.1681970357894897, | |
| "eval_runtime": 0.2494, | |
| "eval_samples_per_second": 874.117, | |
| "eval_steps_per_second": 16.039, | |
| "step": 1392 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "grad_norm": 0.0018475407268851995, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.0001, | |
| "step": 1416 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_f1": 0.8475855910205873, | |
| "eval_loss": 1.1666558980941772, | |
| "eval_runtime": 0.2484, | |
| "eval_samples_per_second": 877.516, | |
| "eval_steps_per_second": 16.101, | |
| "step": 1416 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 0.0015070955269038677, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0001, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_f1": 0.8475855910205873, | |
| "eval_loss": 1.1691235303878784, | |
| "eval_runtime": 0.2503, | |
| "eval_samples_per_second": 870.922, | |
| "eval_steps_per_second": 15.98, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "grad_norm": 0.001129466574639082, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.0001, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_f1": 0.8475855910205873, | |
| "eval_loss": 1.1708507537841797, | |
| "eval_runtime": 0.2483, | |
| "eval_samples_per_second": 878.104, | |
| "eval_steps_per_second": 16.112, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "grad_norm": 0.001501582097262144, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.0001, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_f1": 0.8475855910205873, | |
| "eval_loss": 1.1724032163619995, | |
| "eval_runtime": 0.2483, | |
| "eval_samples_per_second": 877.807, | |
| "eval_steps_per_second": 16.107, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "grad_norm": 0.0010985672706738114, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.0001, | |
| "step": 1512 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_f1": 0.8475855910205873, | |
| "eval_loss": 1.1738693714141846, | |
| "eval_runtime": 0.2494, | |
| "eval_samples_per_second": 874.122, | |
| "eval_steps_per_second": 16.039, | |
| "step": 1512 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "grad_norm": 0.08045843243598938, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.0001, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_f1": 0.8533899487562311, | |
| "eval_loss": 1.1738698482513428, | |
| "eval_runtime": 0.2562, | |
| "eval_samples_per_second": 850.855, | |
| "eval_steps_per_second": 15.612, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 0.002004158915951848, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.0001, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_f1": 0.8471931986968837, | |
| "eval_loss": 1.1787022352218628, | |
| "eval_runtime": 0.2428, | |
| "eval_samples_per_second": 897.779, | |
| "eval_steps_per_second": 16.473, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "grad_norm": 0.001049485057592392, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.0001, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_f1": 0.8471931986968837, | |
| "eval_loss": 1.180769443511963, | |
| "eval_runtime": 0.2525, | |
| "eval_samples_per_second": 863.258, | |
| "eval_steps_per_second": 15.84, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "grad_norm": 0.0010408489033579826, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.0001, | |
| "step": 1608 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_f1": 0.8471931986968837, | |
| "eval_loss": 1.1815507411956787, | |
| "eval_runtime": 0.2505, | |
| "eval_samples_per_second": 870.186, | |
| "eval_steps_per_second": 15.967, | |
| "step": 1608 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "grad_norm": 0.0012064232723787427, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.0001, | |
| "step": 1632 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_f1": 0.8471931986968837, | |
| "eval_loss": 1.182210922241211, | |
| "eval_runtime": 0.2583, | |
| "eval_samples_per_second": 844.028, | |
| "eval_steps_per_second": 15.487, | |
| "step": 1632 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "grad_norm": 0.0014143523294478655, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.0001, | |
| "step": 1656 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_f1": 0.8524668435013261, | |
| "eval_loss": 1.1818993091583252, | |
| "eval_runtime": 0.2452, | |
| "eval_samples_per_second": 888.962, | |
| "eval_steps_per_second": 16.311, | |
| "step": 1656 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2551863638350032.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |