| { | |
| "best_metric": 1.2180595397949219, | |
| "best_model_checkpoint": "./enko_mbartLarge_36p_exp1/checkpoint-25000", | |
| "epoch": 4.180990430177459, | |
| "eval_steps": 5000, | |
| "global_step": 45000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 5e-05, | |
| "loss": 2.3583, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.9844667432974e-05, | |
| "loss": 1.7529, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.9689334865948e-05, | |
| "loss": 1.5715, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.9534002298921995e-05, | |
| "loss": 1.535, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.937866973189599e-05, | |
| "loss": 1.5009, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.9223337164869984e-05, | |
| "loss": 1.4783, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.906800459784399e-05, | |
| "loss": 1.4542, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.8912672030817986e-05, | |
| "loss": 1.441, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.875733946379198e-05, | |
| "loss": 1.425, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.8602006896765975e-05, | |
| "loss": 1.4235, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_bleu": 12.3168, | |
| "eval_gen_len": 14.6634, | |
| "eval_loss": 1.3893214464187622, | |
| "eval_runtime": 1225.507, | |
| "eval_samples_per_second": 17.629, | |
| "eval_steps_per_second": 1.102, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.844667432973998e-05, | |
| "loss": 1.3969, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.829134176271397e-05, | |
| "loss": 1.3901, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.813600919568797e-05, | |
| "loss": 1.3983, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.798067662866197e-05, | |
| "loss": 1.3744, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.782534406163596e-05, | |
| "loss": 1.3712, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.767001149460996e-05, | |
| "loss": 1.3604, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.751467892758396e-05, | |
| "loss": 1.3653, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.7359346360557956e-05, | |
| "loss": 1.3603, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.7204013793531954e-05, | |
| "loss": 1.3188, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.704868122650595e-05, | |
| "loss": 1.3281, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_bleu": 14.3522, | |
| "eval_gen_len": 14.9186, | |
| "eval_loss": 1.2917265892028809, | |
| "eval_runtime": 1232.5734, | |
| "eval_samples_per_second": 17.528, | |
| "eval_steps_per_second": 1.096, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.689334865947995e-05, | |
| "loss": 1.3329, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 4.673801609245395e-05, | |
| "loss": 1.3462, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.6582683525427945e-05, | |
| "loss": 1.3051, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.6427350958401935e-05, | |
| "loss": 1.2672, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.627201839137594e-05, | |
| "loss": 1.2606, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.611668582434994e-05, | |
| "loss": 1.268, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4.596135325732393e-05, | |
| "loss": 1.2733, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 4.5806020690297926e-05, | |
| "loss": 1.2652, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 4.565068812327193e-05, | |
| "loss": 1.2389, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.549535555624592e-05, | |
| "loss": 1.2506, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_bleu": 14.3525, | |
| "eval_gen_len": 14.9494, | |
| "eval_loss": 1.2668566703796387, | |
| "eval_runtime": 1245.6055, | |
| "eval_samples_per_second": 17.345, | |
| "eval_steps_per_second": 1.085, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 4.534002298921992e-05, | |
| "loss": 1.2423, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 4.5184690422193924e-05, | |
| "loss": 1.2214, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 4.5029357855167915e-05, | |
| "loss": 1.1984, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.487402528814191e-05, | |
| "loss": 1.1922, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 4.471869272111591e-05, | |
| "loss": 1.1952, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.456336015408991e-05, | |
| "loss": 1.1998, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 4.4408027587063905e-05, | |
| "loss": 1.1676, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 4.42526950200379e-05, | |
| "loss": 1.175, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 4.40973624530119e-05, | |
| "loss": 1.1674, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 4.39420298859859e-05, | |
| "loss": 1.1603, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_bleu": 15.248, | |
| "eval_gen_len": 15.0062, | |
| "eval_loss": 1.2283018827438354, | |
| "eval_runtime": 1235.7777, | |
| "eval_samples_per_second": 17.483, | |
| "eval_steps_per_second": 1.093, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 4.3786697318959896e-05, | |
| "loss": 1.1944, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.3631364751933894e-05, | |
| "loss": 1.1644, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.347603218490789e-05, | |
| "loss": 1.137, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 4.332069961788189e-05, | |
| "loss": 1.1543, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 4.316536705085589e-05, | |
| "loss": 1.1359, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 4.301003448382988e-05, | |
| "loss": 1.1373, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 4.285470191680388e-05, | |
| "loss": 1.1229, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 4.269936934977788e-05, | |
| "loss": 1.1076, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.254403678275187e-05, | |
| "loss": 1.0849, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.2388704215725875e-05, | |
| "loss": 1.0765, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_bleu": 15.4063, | |
| "eval_gen_len": 14.7808, | |
| "eval_loss": 1.2180595397949219, | |
| "eval_runtime": 1212.0146, | |
| "eval_samples_per_second": 17.826, | |
| "eval_steps_per_second": 1.115, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.223337164869987e-05, | |
| "loss": 1.107, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 4.2078039081673864e-05, | |
| "loss": 1.1183, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 4.192270651464786e-05, | |
| "loss": 1.1004, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 4.1767373947621866e-05, | |
| "loss": 1.0793, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.161204138059586e-05, | |
| "loss": 1.0901, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.1456708813569854e-05, | |
| "loss": 1.1028, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 4.130137624654385e-05, | |
| "loss": 1.1398, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 4.114604367951785e-05, | |
| "loss": 1.088, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 4.099071111249185e-05, | |
| "loss": 1.0459, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 4.0835378545465845e-05, | |
| "loss": 1.1019, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_bleu": 14.3608, | |
| "eval_gen_len": 14.9014, | |
| "eval_loss": 1.2753331661224365, | |
| "eval_runtime": 1243.9853, | |
| "eval_samples_per_second": 17.368, | |
| "eval_steps_per_second": 1.086, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 4.068004597843984e-05, | |
| "loss": 1.0986, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 4.052471341141384e-05, | |
| "loss": 1.107, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.036938084438784e-05, | |
| "loss": 1.1234, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 4.021404827736183e-05, | |
| "loss": 1.138, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 4.0058715710335834e-05, | |
| "loss": 1.1106, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 3.990338314330983e-05, | |
| "loss": 1.0363, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 3.974805057628382e-05, | |
| "loss": 1.01, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 3.959271800925782e-05, | |
| "loss": 1.0107, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3.9437385442231824e-05, | |
| "loss": 1.0122, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 3.9282052875205815e-05, | |
| "loss": 1.0504, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_bleu": 15.3253, | |
| "eval_gen_len": 14.7948, | |
| "eval_loss": 1.2334309816360474, | |
| "eval_runtime": 1248.5109, | |
| "eval_samples_per_second": 17.305, | |
| "eval_steps_per_second": 1.082, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 3.912672030817981e-05, | |
| "loss": 1.0407, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 3.897138774115382e-05, | |
| "loss": 1.0186, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 3.881605517412781e-05, | |
| "loss": 0.9753, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 3.8660722607101806e-05, | |
| "loss": 0.9714, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 3.8505390040075804e-05, | |
| "loss": 0.9734, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 3.83500574730498e-05, | |
| "loss": 0.963, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 3.81947249060238e-05, | |
| "loss": 1.0455, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 3.80393923389978e-05, | |
| "loss": 1.064, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 3.7884059771971794e-05, | |
| "loss": 1.005, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 3.772872720494579e-05, | |
| "loss": 0.9431, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "eval_bleu": 15.2534, | |
| "eval_gen_len": 14.7293, | |
| "eval_loss": 1.2512198686599731, | |
| "eval_runtime": 1212.804, | |
| "eval_samples_per_second": 17.814, | |
| "eval_steps_per_second": 1.114, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 3.757339463791979e-05, | |
| "loss": 0.8958, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 3.741806207089378e-05, | |
| "loss": 0.865, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 3.7262729503867785e-05, | |
| "loss": 0.8829, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 3.710739693684178e-05, | |
| "loss": 0.9097, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 3.6952064369815774e-05, | |
| "loss": 0.8159, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.679673180278977e-05, | |
| "loss": 0.8056, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 3.6641399235763776e-05, | |
| "loss": 0.7523, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 3.648606666873777e-05, | |
| "loss": 0.7307, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.6330734101711764e-05, | |
| "loss": 0.7998, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 3.617540153468577e-05, | |
| "loss": 0.8394, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "eval_bleu": 14.9999, | |
| "eval_gen_len": 14.7993, | |
| "eval_loss": 1.2971030473709106, | |
| "eval_runtime": 1213.4816, | |
| "eval_samples_per_second": 17.804, | |
| "eval_steps_per_second": 1.113, | |
| "step": 45000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 161445, | |
| "num_train_epochs": 15, | |
| "save_steps": 5000, | |
| "total_flos": 1.5605414779085128e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |