| { | |
| "best_metric": 0.8995155096054077, | |
| "best_model_checkpoint": "./mbartLarge_koja_37p/checkpoint-33000", | |
| "epoch": 4.844107803417298, | |
| "eval_steps": 5500, | |
| "global_step": 55000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 3.5939, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 3.0254, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9133, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.9936851145520225e-05, | |
| "loss": 1.672, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.987370229104044e-05, | |
| "loss": 1.56, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.9810553436560664e-05, | |
| "loss": 1.4874, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.974740458208088e-05, | |
| "loss": 1.4345, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.96842557276011e-05, | |
| "loss": 1.3676, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.9621106873121326e-05, | |
| "loss": 1.3183, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.955795801864154e-05, | |
| "loss": 1.2987, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.9494809164161765e-05, | |
| "loss": 1.2712, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_bleu": 3.0858, | |
| "eval_gen_len": 18.1085, | |
| "eval_loss": 1.2203737497329712, | |
| "eval_runtime": 1528.8304, | |
| "eval_samples_per_second": 14.853, | |
| "eval_steps_per_second": 0.929, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.943166030968199e-05, | |
| "loss": 1.2405, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.9368511455202204e-05, | |
| "loss": 1.2262, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.930536260072243e-05, | |
| "loss": 1.1959, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.924221374624264e-05, | |
| "loss": 1.182, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.9179064891762866e-05, | |
| "loss": 1.1683, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.911591603728309e-05, | |
| "loss": 1.1489, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.9052767182803305e-05, | |
| "loss": 1.1365, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.898961832832353e-05, | |
| "loss": 1.1313, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.8926469473843744e-05, | |
| "loss": 1.1172, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.886332061936397e-05, | |
| "loss": 1.1123, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.880017176488419e-05, | |
| "loss": 1.0946, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_bleu": 3.3162, | |
| "eval_gen_len": 17.8651, | |
| "eval_loss": 1.0577712059020996, | |
| "eval_runtime": 1538.4057, | |
| "eval_samples_per_second": 14.76, | |
| "eval_steps_per_second": 0.923, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 4.8737022910404406e-05, | |
| "loss": 1.0856, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 4.867387405592463e-05, | |
| "loss": 1.0716, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.8610725201444845e-05, | |
| "loss": 1.0736, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.854757634696507e-05, | |
| "loss": 1.0367, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.848442749248529e-05, | |
| "loss": 1.0204, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.842127863800551e-05, | |
| "loss": 1.0533, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.835812978352573e-05, | |
| "loss": 1.0426, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.829498092904595e-05, | |
| "loss": 1.0261, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.823183207456617e-05, | |
| "loss": 1.0111, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 4.816868322008639e-05, | |
| "loss": 0.9862, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.810553436560661e-05, | |
| "loss": 0.9546, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_bleu": 5.6024, | |
| "eval_gen_len": 17.902, | |
| "eval_loss": 0.9687988758087158, | |
| "eval_runtime": 1530.7999, | |
| "eval_samples_per_second": 14.833, | |
| "eval_steps_per_second": 0.928, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.804238551112683e-05, | |
| "loss": 0.9642, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.7979236656647054e-05, | |
| "loss": 0.9364, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 4.791608780216727e-05, | |
| "loss": 0.931, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 4.785293894768749e-05, | |
| "loss": 0.915, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.778979009320771e-05, | |
| "loss": 0.9043, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 4.772664123872793e-05, | |
| "loss": 0.9158, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 4.7663492384248155e-05, | |
| "loss": 0.9068, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 4.760034352976837e-05, | |
| "loss": 0.9314, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.7537194675288594e-05, | |
| "loss": 0.9112, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 4.747404582080881e-05, | |
| "loss": 0.8862, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 4.741089696632903e-05, | |
| "loss": 0.89, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_bleu": 5.1453, | |
| "eval_gen_len": 17.6144, | |
| "eval_loss": 0.941440761089325, | |
| "eval_runtime": 1490.4507, | |
| "eval_samples_per_second": 15.235, | |
| "eval_steps_per_second": 0.953, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 4.7347748111849256e-05, | |
| "loss": 0.8726, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.728459925736947e-05, | |
| "loss": 0.8749, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 4.7221450402889695e-05, | |
| "loss": 0.858, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.715830154840992e-05, | |
| "loss": 0.8479, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 4.7095152693930134e-05, | |
| "loss": 0.8394, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 4.703200383945036e-05, | |
| "loss": 0.8472, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 4.696885498497057e-05, | |
| "loss": 0.7793, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.6905706130490796e-05, | |
| "loss": 0.7753, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.684255727601102e-05, | |
| "loss": 0.7655, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.6779408421531235e-05, | |
| "loss": 0.7666, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 4.671625956705146e-05, | |
| "loss": 0.834, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_bleu": 5.3985, | |
| "eval_gen_len": 17.6899, | |
| "eval_loss": 0.9212985038757324, | |
| "eval_runtime": 1489.7985, | |
| "eval_samples_per_second": 15.242, | |
| "eval_steps_per_second": 0.953, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 4.6653110712571674e-05, | |
| "loss": 0.8098, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 4.65899618580919e-05, | |
| "loss": 0.8153, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.652681300361212e-05, | |
| "loss": 0.7837, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.6463664149132336e-05, | |
| "loss": 0.8255, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 4.640051529465256e-05, | |
| "loss": 0.8345, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 4.6337366440172775e-05, | |
| "loss": 0.8061, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 4.6274217585693e-05, | |
| "loss": 0.7743, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 4.621106873121322e-05, | |
| "loss": 0.7759, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 4.614791987673344e-05, | |
| "loss": 0.7777, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 4.608477102225366e-05, | |
| "loss": 0.7537, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 4.602162216777388e-05, | |
| "loss": 0.7439, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_bleu": 6.2934, | |
| "eval_gen_len": 17.4862, | |
| "eval_loss": 0.8995155096054077, | |
| "eval_runtime": 1459.4956, | |
| "eval_samples_per_second": 15.558, | |
| "eval_steps_per_second": 0.973, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 4.59584733132941e-05, | |
| "loss": 0.7424, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.589532445881432e-05, | |
| "loss": 0.7416, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.583217560433454e-05, | |
| "loss": 0.7722, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 4.576902674985476e-05, | |
| "loss": 0.7967, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.5705877895374984e-05, | |
| "loss": 0.7323, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 4.56427290408952e-05, | |
| "loss": 0.7171, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 4.557958018641542e-05, | |
| "loss": 0.7044, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 4.551643133193564e-05, | |
| "loss": 0.6828, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 4.545328247745586e-05, | |
| "loss": 0.7149, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 4.5390133622976085e-05, | |
| "loss": 0.6622, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 4.53269847684963e-05, | |
| "loss": 0.6803, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "eval_bleu": 6.3565, | |
| "eval_gen_len": 17.8899, | |
| "eval_loss": 0.9016386270523071, | |
| "eval_runtime": 1493.3684, | |
| "eval_samples_per_second": 15.205, | |
| "eval_steps_per_second": 0.951, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 4.5263835914016524e-05, | |
| "loss": 0.7079, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 4.520068705953675e-05, | |
| "loss": 0.6978, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 4.513753820505696e-05, | |
| "loss": 0.7528, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 4.5074389350577186e-05, | |
| "loss": 0.7558, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 4.50112404960974e-05, | |
| "loss": 0.7742, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 4.4948091641617625e-05, | |
| "loss": 0.7505, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 4.488494278713785e-05, | |
| "loss": 0.7185, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 4.4821793932658064e-05, | |
| "loss": 0.7024, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 4.475864507817829e-05, | |
| "loss": 0.7262, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 4.46954962236985e-05, | |
| "loss": 0.7394, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 4.4632347369218726e-05, | |
| "loss": 0.733, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "eval_bleu": 7.0351, | |
| "eval_gen_len": 17.6112, | |
| "eval_loss": 0.9225591421127319, | |
| "eval_runtime": 1483.9686, | |
| "eval_samples_per_second": 15.302, | |
| "eval_steps_per_second": 0.957, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 4.456919851473895e-05, | |
| "loss": 0.742, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 4.4506049660259165e-05, | |
| "loss": 0.735, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 4.444290080577939e-05, | |
| "loss": 0.6992, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 4.4379751951299604e-05, | |
| "loss": 0.7007, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 4.431660309681983e-05, | |
| "loss": 0.6979, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 4.425345424234005e-05, | |
| "loss": 0.6515, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 4.4190305387860266e-05, | |
| "loss": 0.6919, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 4.412715653338049e-05, | |
| "loss": 0.6241, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 4.406400767890071e-05, | |
| "loss": 0.6196, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 4.400085882442093e-05, | |
| "loss": 0.6557, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 4.393770996994115e-05, | |
| "loss": 0.6601, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "eval_bleu": 5.3084, | |
| "eval_gen_len": 17.4292, | |
| "eval_loss": 0.9806875586509705, | |
| "eval_runtime": 1461.811, | |
| "eval_samples_per_second": 15.533, | |
| "eval_steps_per_second": 0.971, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 4.387456111546137e-05, | |
| "loss": 0.7354, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 4.381141226098159e-05, | |
| "loss": 0.7452, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 4.374826340650181e-05, | |
| "loss": 0.7175, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 4.368511455202203e-05, | |
| "loss": 0.7101, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 4.362196569754225e-05, | |
| "loss": 0.7068, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 4.355881684306247e-05, | |
| "loss": 0.6404, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 4.349566798858269e-05, | |
| "loss": 0.6316, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 4.3432519134102913e-05, | |
| "loss": 0.598, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.336937027962313e-05, | |
| "loss": 0.6091, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 4.330622142514335e-05, | |
| "loss": 0.695, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 4.324307257066357e-05, | |
| "loss": 0.6933, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "eval_bleu": 6.8389, | |
| "eval_gen_len": 17.5131, | |
| "eval_loss": 0.923828661441803, | |
| "eval_runtime": 1455.6892, | |
| "eval_samples_per_second": 15.599, | |
| "eval_steps_per_second": 0.975, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "step": 55000, | |
| "total_flos": 1.907212892088828e+18, | |
| "train_loss": 0.9352452936345881, | |
| "train_runtime": 56979.8243, | |
| "train_samples_per_second": 111.586, | |
| "train_steps_per_second": 6.974 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 397390, | |
| "num_train_epochs": 35, | |
| "save_steps": 5500, | |
| "total_flos": 1.907212892088828e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |