| { | |
| "best_metric": 0.6427481174468994, | |
| "best_model_checkpoint": "./models/adapters_mlm_cn/mt/checkpoint-40000", | |
| "epoch": 82.81573498964804, | |
| "eval_steps": 500, | |
| "global_step": 40000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 2.6689, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_accuracy": 0.6676783004552352, | |
| "eval_loss": 2.1344573497772217, | |
| "eval_runtime": 1.6941, | |
| "eval_samples_per_second": 506.449, | |
| "eval_steps_per_second": 31.874, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 4.9e-05, | |
| "loss": 2.1415, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_accuracy": 0.6926470588235294, | |
| "eval_loss": 1.8435733318328857, | |
| "eval_runtime": 1.6896, | |
| "eval_samples_per_second": 507.824, | |
| "eval_steps_per_second": 31.961, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 4.85e-05, | |
| "loss": 1.9421, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "eval_accuracy": 0.690674753601213, | |
| "eval_loss": 1.787391185760498, | |
| "eval_runtime": 1.6956, | |
| "eval_samples_per_second": 506.016, | |
| "eval_steps_per_second": 31.847, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.7588, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_accuracy": 0.7012509197939661, | |
| "eval_loss": 1.760498285293579, | |
| "eval_runtime": 1.6894, | |
| "eval_samples_per_second": 507.882, | |
| "eval_steps_per_second": 31.965, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.6729, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "eval_accuracy": 0.6956845238095238, | |
| "eval_loss": 1.7567747831344604, | |
| "eval_runtime": 1.6937, | |
| "eval_samples_per_second": 506.597, | |
| "eval_steps_per_second": 31.884, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.596, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 1.5006115436553955, | |
| "eval_runtime": 1.6969, | |
| "eval_samples_per_second": 505.627, | |
| "eval_steps_per_second": 31.823, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 1.5778, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "eval_accuracy": 0.7450832072617246, | |
| "eval_loss": 1.3923866748809814, | |
| "eval_runtime": 1.6898, | |
| "eval_samples_per_second": 507.75, | |
| "eval_steps_per_second": 31.956, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.4821, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "eval_accuracy": 0.7099236641221374, | |
| "eval_loss": 1.609680414199829, | |
| "eval_runtime": 1.6898, | |
| "eval_samples_per_second": 507.758, | |
| "eval_steps_per_second": 31.957, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 4.55e-05, | |
| "loss": 1.4183, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "eval_accuracy": 0.7490551776266062, | |
| "eval_loss": 1.3551626205444336, | |
| "eval_runtime": 1.6905, | |
| "eval_samples_per_second": 507.539, | |
| "eval_steps_per_second": 31.943, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.35, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.4197, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.35, | |
| "eval_accuracy": 0.7513471901462664, | |
| "eval_loss": 1.284741997718811, | |
| "eval_runtime": 1.6878, | |
| "eval_samples_per_second": 508.353, | |
| "eval_steps_per_second": 31.994, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 1.3156, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "eval_accuracy": 0.7496318114874816, | |
| "eval_loss": 1.3172950744628906, | |
| "eval_runtime": 1.6891, | |
| "eval_samples_per_second": 507.956, | |
| "eval_steps_per_second": 31.969, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.2882, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "eval_accuracy": 0.7738095238095238, | |
| "eval_loss": 1.2816879749298096, | |
| "eval_runtime": 1.6955, | |
| "eval_samples_per_second": 506.058, | |
| "eval_steps_per_second": 31.85, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "learning_rate": 4.35e-05, | |
| "loss": 1.2692, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "eval_accuracy": 0.775112443778111, | |
| "eval_loss": 1.189226746559143, | |
| "eval_runtime": 1.6876, | |
| "eval_samples_per_second": 508.403, | |
| "eval_steps_per_second": 31.997, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 4.3e-05, | |
| "loss": 1.2368, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "eval_accuracy": 0.7816432272390822, | |
| "eval_loss": 1.2362936735153198, | |
| "eval_runtime": 1.6828, | |
| "eval_samples_per_second": 509.861, | |
| "eval_steps_per_second": 32.089, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 15.53, | |
| "learning_rate": 4.25e-05, | |
| "loss": 1.1975, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 15.53, | |
| "eval_accuracy": 0.76996336996337, | |
| "eval_loss": 1.2442289590835571, | |
| "eval_runtime": 1.6849, | |
| "eval_samples_per_second": 509.233, | |
| "eval_steps_per_second": 32.05, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.1907, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 16.56, | |
| "eval_accuracy": 0.7720320466132556, | |
| "eval_loss": 1.256901502609253, | |
| "eval_runtime": 1.6884, | |
| "eval_samples_per_second": 508.188, | |
| "eval_steps_per_second": 31.984, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 4.15e-05, | |
| "loss": 1.1231, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "eval_accuracy": 0.776085300837776, | |
| "eval_loss": 1.13861083984375, | |
| "eval_runtime": 1.6842, | |
| "eval_samples_per_second": 509.436, | |
| "eval_steps_per_second": 32.062, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 18.63, | |
| "learning_rate": 4.1e-05, | |
| "loss": 1.0873, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 18.63, | |
| "eval_accuracy": 0.7855547282204021, | |
| "eval_loss": 1.2104856967926025, | |
| "eval_runtime": 1.6846, | |
| "eval_samples_per_second": 509.309, | |
| "eval_steps_per_second": 32.054, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 19.67, | |
| "learning_rate": 4.05e-05, | |
| "loss": 1.1242, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 19.67, | |
| "eval_accuracy": 0.7737909516380655, | |
| "eval_loss": 1.214229702949524, | |
| "eval_runtime": 1.6848, | |
| "eval_samples_per_second": 509.258, | |
| "eval_steps_per_second": 32.051, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 20.7, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0367, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 20.7, | |
| "eval_accuracy": 0.7712369597615499, | |
| "eval_loss": 1.2120734453201294, | |
| "eval_runtime": 1.6938, | |
| "eval_samples_per_second": 506.56, | |
| "eval_steps_per_second": 31.881, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 1.0869, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "eval_accuracy": 0.7955390334572491, | |
| "eval_loss": 1.0782362222671509, | |
| "eval_runtime": 1.6863, | |
| "eval_samples_per_second": 508.795, | |
| "eval_steps_per_second": 32.022, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 22.77, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 1.0353, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 22.77, | |
| "eval_accuracy": 0.8068535825545171, | |
| "eval_loss": 0.9917858839035034, | |
| "eval_runtime": 1.6841, | |
| "eval_samples_per_second": 509.483, | |
| "eval_steps_per_second": 32.065, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 23.81, | |
| "learning_rate": 3.85e-05, | |
| "loss": 1.0324, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 23.81, | |
| "eval_accuracy": 0.7971233913701741, | |
| "eval_loss": 1.0908266305923462, | |
| "eval_runtime": 1.6848, | |
| "eval_samples_per_second": 509.246, | |
| "eval_steps_per_second": 32.05, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 24.84, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.0145, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 24.84, | |
| "eval_accuracy": 0.7975460122699386, | |
| "eval_loss": 1.0944875478744507, | |
| "eval_runtime": 1.6827, | |
| "eval_samples_per_second": 509.9, | |
| "eval_steps_per_second": 32.092, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 25.88, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.9951, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 25.88, | |
| "eval_accuracy": 0.8028064992614475, | |
| "eval_loss": 1.000519037246704, | |
| "eval_runtime": 1.6933, | |
| "eval_samples_per_second": 506.714, | |
| "eval_steps_per_second": 31.891, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.9483, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "eval_accuracy": 0.8186646433990895, | |
| "eval_loss": 0.963790237903595, | |
| "eval_runtime": 1.6874, | |
| "eval_samples_per_second": 508.479, | |
| "eval_steps_per_second": 32.002, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 27.95, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.9304, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 27.95, | |
| "eval_accuracy": 0.8204747774480712, | |
| "eval_loss": 0.9761123657226562, | |
| "eval_runtime": 1.6869, | |
| "eval_samples_per_second": 508.622, | |
| "eval_steps_per_second": 32.011, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.8835, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "eval_accuracy": 0.8045801526717558, | |
| "eval_loss": 1.062032699584961, | |
| "eval_runtime": 1.6883, | |
| "eval_samples_per_second": 508.21, | |
| "eval_steps_per_second": 31.985, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 30.02, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.9097, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 30.02, | |
| "eval_accuracy": 0.806015037593985, | |
| "eval_loss": 0.9137569069862366, | |
| "eval_runtime": 1.6924, | |
| "eval_samples_per_second": 506.97, | |
| "eval_steps_per_second": 31.907, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 31.06, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.9293, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 31.06, | |
| "eval_accuracy": 0.8176197836166924, | |
| "eval_loss": 0.918023943901062, | |
| "eval_runtime": 1.6905, | |
| "eval_samples_per_second": 507.53, | |
| "eval_steps_per_second": 31.942, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 32.09, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.9043, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 32.09, | |
| "eval_accuracy": 0.8208269525267994, | |
| "eval_loss": 0.9214709401130676, | |
| "eval_runtime": 1.691, | |
| "eval_samples_per_second": 507.403, | |
| "eval_steps_per_second": 31.934, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 33.13, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.8581, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 33.13, | |
| "eval_accuracy": 0.822452229299363, | |
| "eval_loss": 0.9624596834182739, | |
| "eval_runtime": 1.6897, | |
| "eval_samples_per_second": 507.793, | |
| "eval_steps_per_second": 31.959, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 34.16, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.8638, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 34.16, | |
| "eval_accuracy": 0.8367816091954023, | |
| "eval_loss": 0.8585591316223145, | |
| "eval_runtime": 1.6912, | |
| "eval_samples_per_second": 507.329, | |
| "eval_steps_per_second": 31.93, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 35.2, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.874, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 35.2, | |
| "eval_accuracy": 0.8135072908672295, | |
| "eval_loss": 1.0043973922729492, | |
| "eval_runtime": 1.6896, | |
| "eval_samples_per_second": 507.801, | |
| "eval_steps_per_second": 31.96, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 36.23, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.8235, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 36.23, | |
| "eval_accuracy": 0.8183890577507599, | |
| "eval_loss": 0.9755066633224487, | |
| "eval_runtime": 1.6947, | |
| "eval_samples_per_second": 506.289, | |
| "eval_steps_per_second": 31.864, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 37.27, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.8589, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 37.27, | |
| "eval_accuracy": 0.8291761148904006, | |
| "eval_loss": 0.9042153358459473, | |
| "eval_runtime": 1.6905, | |
| "eval_samples_per_second": 507.55, | |
| "eval_steps_per_second": 31.944, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 38.3, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.8107, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 38.3, | |
| "eval_accuracy": 0.8272327964860908, | |
| "eval_loss": 0.8821109533309937, | |
| "eval_runtime": 1.6895, | |
| "eval_samples_per_second": 507.845, | |
| "eval_steps_per_second": 31.962, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 39.34, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.8346, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 39.34, | |
| "eval_accuracy": 0.8248286367098249, | |
| "eval_loss": 0.9061236381530762, | |
| "eval_runtime": 1.6919, | |
| "eval_samples_per_second": 507.136, | |
| "eval_steps_per_second": 31.918, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 40.37, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.8393, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 40.37, | |
| "eval_accuracy": 0.8234854151084517, | |
| "eval_loss": 0.9795840978622437, | |
| "eval_runtime": 1.6939, | |
| "eval_samples_per_second": 506.513, | |
| "eval_steps_per_second": 31.878, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 41.41, | |
| "learning_rate": 3e-05, | |
| "loss": 0.789, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 41.41, | |
| "eval_accuracy": 0.833076923076923, | |
| "eval_loss": 0.9014851450920105, | |
| "eval_runtime": 1.689, | |
| "eval_samples_per_second": 508.0, | |
| "eval_steps_per_second": 31.972, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 42.44, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.8121, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 42.44, | |
| "eval_accuracy": 0.8385913426265591, | |
| "eval_loss": 0.8589309453964233, | |
| "eval_runtime": 1.6873, | |
| "eval_samples_per_second": 508.516, | |
| "eval_steps_per_second": 32.005, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 43.48, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.7709, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 43.48, | |
| "eval_accuracy": 0.8350903614457831, | |
| "eval_loss": 0.8835715055465698, | |
| "eval_runtime": 1.6829, | |
| "eval_samples_per_second": 509.835, | |
| "eval_steps_per_second": 32.088, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 44.51, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.7922, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 44.51, | |
| "eval_accuracy": 0.817974105102818, | |
| "eval_loss": 0.9523779153823853, | |
| "eval_runtime": 1.6863, | |
| "eval_samples_per_second": 508.799, | |
| "eval_steps_per_second": 32.022, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 45.55, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.7457, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 45.55, | |
| "eval_accuracy": 0.8364451082897685, | |
| "eval_loss": 0.8350428938865662, | |
| "eval_runtime": 1.6901, | |
| "eval_samples_per_second": 507.673, | |
| "eval_steps_per_second": 31.951, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 46.58, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.7386, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 46.58, | |
| "eval_accuracy": 0.8340807174887892, | |
| "eval_loss": 0.9024766087532043, | |
| "eval_runtime": 1.6912, | |
| "eval_samples_per_second": 507.34, | |
| "eval_steps_per_second": 31.93, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 47.62, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.7515, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 47.62, | |
| "eval_accuracy": 0.8390166534496432, | |
| "eval_loss": 0.9091906547546387, | |
| "eval_runtime": 1.686, | |
| "eval_samples_per_second": 508.899, | |
| "eval_steps_per_second": 32.029, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 48.65, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.7324, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 48.65, | |
| "eval_accuracy": 0.8420647149460708, | |
| "eval_loss": 0.8322407007217407, | |
| "eval_runtime": 1.6918, | |
| "eval_samples_per_second": 507.153, | |
| "eval_steps_per_second": 31.919, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 49.69, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.7314, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 49.69, | |
| "eval_accuracy": 0.8477078477078477, | |
| "eval_loss": 0.7967829704284668, | |
| "eval_runtime": 1.6933, | |
| "eval_samples_per_second": 506.713, | |
| "eval_steps_per_second": 31.891, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 50.72, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.7442, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 50.72, | |
| "eval_accuracy": 0.8324407039020658, | |
| "eval_loss": 0.930473268032074, | |
| "eval_runtime": 1.6828, | |
| "eval_samples_per_second": 509.873, | |
| "eval_steps_per_second": 32.09, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 51.76, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7074, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 51.76, | |
| "eval_accuracy": 0.820839580209895, | |
| "eval_loss": 1.001060962677002, | |
| "eval_runtime": 1.6867, | |
| "eval_samples_per_second": 508.672, | |
| "eval_steps_per_second": 32.014, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 52.8, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.739, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 52.8, | |
| "eval_accuracy": 0.8330945558739254, | |
| "eval_loss": 0.8732258677482605, | |
| "eval_runtime": 1.6896, | |
| "eval_samples_per_second": 507.823, | |
| "eval_steps_per_second": 31.961, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 53.83, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.7243, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 53.83, | |
| "eval_accuracy": 0.8479880774962743, | |
| "eval_loss": 0.7857112288475037, | |
| "eval_runtime": 1.687, | |
| "eval_samples_per_second": 508.591, | |
| "eval_steps_per_second": 32.009, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 54.87, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.6842, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 54.87, | |
| "eval_accuracy": 0.8377192982456141, | |
| "eval_loss": 0.7945135235786438, | |
| "eval_runtime": 1.6902, | |
| "eval_samples_per_second": 507.642, | |
| "eval_steps_per_second": 31.949, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 55.9, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.6991, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 55.9, | |
| "eval_accuracy": 0.8275351591413768, | |
| "eval_loss": 0.9627696871757507, | |
| "eval_runtime": 1.6871, | |
| "eval_samples_per_second": 508.578, | |
| "eval_steps_per_second": 32.008, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 56.94, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.6896, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 56.94, | |
| "eval_accuracy": 0.840960240060015, | |
| "eval_loss": 0.8363039493560791, | |
| "eval_runtime": 1.684, | |
| "eval_samples_per_second": 509.495, | |
| "eval_steps_per_second": 32.066, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 57.97, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.6925, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 57.97, | |
| "eval_accuracy": 0.8391812865497076, | |
| "eval_loss": 0.8432921767234802, | |
| "eval_runtime": 1.6968, | |
| "eval_samples_per_second": 505.655, | |
| "eval_steps_per_second": 31.824, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 59.01, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.7081, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 59.01, | |
| "eval_accuracy": 0.8223048327137547, | |
| "eval_loss": 1.0085676908493042, | |
| "eval_runtime": 1.69, | |
| "eval_samples_per_second": 507.688, | |
| "eval_steps_per_second": 31.952, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 60.04, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.6598, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 60.04, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.9250668883323669, | |
| "eval_runtime": 1.686, | |
| "eval_samples_per_second": 508.895, | |
| "eval_steps_per_second": 32.028, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 61.08, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.6677, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 61.08, | |
| "eval_accuracy": 0.8437047756874095, | |
| "eval_loss": 0.8822752237319946, | |
| "eval_runtime": 1.693, | |
| "eval_samples_per_second": 506.807, | |
| "eval_steps_per_second": 31.897, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 62.11, | |
| "learning_rate": 2e-05, | |
| "loss": 0.695, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 62.11, | |
| "eval_accuracy": 0.8560371517027864, | |
| "eval_loss": 0.7750544548034668, | |
| "eval_runtime": 1.6969, | |
| "eval_samples_per_second": 505.632, | |
| "eval_steps_per_second": 31.823, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 63.15, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.7108, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 63.15, | |
| "eval_accuracy": 0.8481104651162791, | |
| "eval_loss": 0.8452057242393494, | |
| "eval_runtime": 1.6974, | |
| "eval_samples_per_second": 505.49, | |
| "eval_steps_per_second": 31.814, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 64.18, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.6721, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 64.18, | |
| "eval_accuracy": 0.8413284132841329, | |
| "eval_loss": 0.8559600114822388, | |
| "eval_runtime": 1.6936, | |
| "eval_samples_per_second": 506.623, | |
| "eval_steps_per_second": 31.885, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 65.22, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.6571, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 65.22, | |
| "eval_accuracy": 0.8163109756097561, | |
| "eval_loss": 0.98003089427948, | |
| "eval_runtime": 1.6913, | |
| "eval_samples_per_second": 507.303, | |
| "eval_steps_per_second": 31.928, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 66.25, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.6891, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 66.25, | |
| "eval_accuracy": 0.8457446808510638, | |
| "eval_loss": 0.8105884194374084, | |
| "eval_runtime": 1.6942, | |
| "eval_samples_per_second": 506.435, | |
| "eval_steps_per_second": 31.874, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 67.29, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.6541, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 67.29, | |
| "eval_accuracy": 0.8429752066115702, | |
| "eval_loss": 0.8197007179260254, | |
| "eval_runtime": 1.6912, | |
| "eval_samples_per_second": 507.332, | |
| "eval_steps_per_second": 31.93, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 68.32, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.6559, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 68.32, | |
| "eval_accuracy": 0.8388305847076462, | |
| "eval_loss": 0.8678442239761353, | |
| "eval_runtime": 1.6945, | |
| "eval_samples_per_second": 506.35, | |
| "eval_steps_per_second": 31.868, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 69.36, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.6554, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 69.36, | |
| "eval_accuracy": 0.8661764705882353, | |
| "eval_loss": 0.7396097183227539, | |
| "eval_runtime": 1.6934, | |
| "eval_samples_per_second": 506.658, | |
| "eval_steps_per_second": 31.888, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 70.39, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.618, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 70.39, | |
| "eval_accuracy": 0.8375634517766497, | |
| "eval_loss": 0.8517589569091797, | |
| "eval_runtime": 1.6983, | |
| "eval_samples_per_second": 505.224, | |
| "eval_steps_per_second": 31.797, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 71.43, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.6558, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 71.43, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.7705618739128113, | |
| "eval_runtime": 1.6954, | |
| "eval_samples_per_second": 506.065, | |
| "eval_steps_per_second": 31.85, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 72.46, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.6034, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 72.46, | |
| "eval_accuracy": 0.8517699115044248, | |
| "eval_loss": 0.7829406261444092, | |
| "eval_runtime": 1.6974, | |
| "eval_samples_per_second": 505.471, | |
| "eval_steps_per_second": 31.813, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 73.5, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.6336, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 73.5, | |
| "eval_accuracy": 0.8591445427728613, | |
| "eval_loss": 0.7834987640380859, | |
| "eval_runtime": 1.6914, | |
| "eval_samples_per_second": 507.26, | |
| "eval_steps_per_second": 31.925, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 74.53, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.6287, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 74.53, | |
| "eval_accuracy": 0.8574748257164988, | |
| "eval_loss": 0.7547706961631775, | |
| "eval_runtime": 1.6906, | |
| "eval_samples_per_second": 507.513, | |
| "eval_steps_per_second": 31.941, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 75.57, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.6065, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 75.57, | |
| "eval_accuracy": 0.8508005822416302, | |
| "eval_loss": 0.8541703224182129, | |
| "eval_runtime": 1.6919, | |
| "eval_samples_per_second": 507.134, | |
| "eval_steps_per_second": 31.918, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 76.6, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.6029, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 76.6, | |
| "eval_accuracy": 0.8405267008046818, | |
| "eval_loss": 0.8202521800994873, | |
| "eval_runtime": 1.6903, | |
| "eval_samples_per_second": 507.595, | |
| "eval_steps_per_second": 31.947, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 77.64, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.6208, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 77.64, | |
| "eval_accuracy": 0.8661417322834646, | |
| "eval_loss": 0.7082335948944092, | |
| "eval_runtime": 1.6867, | |
| "eval_samples_per_second": 508.681, | |
| "eval_steps_per_second": 32.015, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 78.67, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.64, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 78.67, | |
| "eval_accuracy": 0.8410295230885693, | |
| "eval_loss": 0.8504825234413147, | |
| "eval_runtime": 1.6943, | |
| "eval_samples_per_second": 506.417, | |
| "eval_steps_per_second": 31.872, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 79.71, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.6144, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 79.71, | |
| "eval_accuracy": 0.8603636363636363, | |
| "eval_loss": 0.7246142625808716, | |
| "eval_runtime": 1.6864, | |
| "eval_samples_per_second": 508.77, | |
| "eval_steps_per_second": 32.02, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 80.75, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.6507, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 80.75, | |
| "eval_accuracy": 0.861132660977502, | |
| "eval_loss": 0.7150202393531799, | |
| "eval_runtime": 1.701, | |
| "eval_samples_per_second": 504.398, | |
| "eval_steps_per_second": 31.745, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 81.78, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.6177, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 81.78, | |
| "eval_accuracy": 0.84, | |
| "eval_loss": 0.9331970810890198, | |
| "eval_runtime": 1.6939, | |
| "eval_samples_per_second": 506.536, | |
| "eval_steps_per_second": 31.88, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 82.82, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6159, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 82.82, | |
| "eval_accuracy": 0.8733488733488733, | |
| "eval_loss": 0.6427481174468994, | |
| "eval_runtime": 1.6965, | |
| "eval_samples_per_second": 505.755, | |
| "eval_steps_per_second": 31.831, | |
| "step": 40000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 50000, | |
| "num_train_epochs": 104, | |
| "save_steps": 500, | |
| "total_flos": 6042662847119360.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |