| { | |
| "best_metric": 0.8551470588235294, | |
| "best_model_checkpoint": "retrieval_model/run-3/checkpoint-35612", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 35612, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.4704519236458057e-06, | |
| "loss": 0.4686, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 3.4210321821949366e-06, | |
| "loss": 0.3929, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.3716124407440683e-06, | |
| "loss": 0.3526, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.3221926992931996e-06, | |
| "loss": 0.3448, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.272772957842331e-06, | |
| "loss": 0.3648, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 3.223353216391462e-06, | |
| "loss": 0.3417, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.173933474940593e-06, | |
| "loss": 0.3259, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.124513733489725e-06, | |
| "loss": 0.3513, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.0750939920388557e-06, | |
| "loss": 0.3343, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3.025674250587987e-06, | |
| "loss": 0.3465, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.9762545091371184e-06, | |
| "loss": 0.3218, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 2.9268347676862497e-06, | |
| "loss": 0.3134, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 2.877415026235381e-06, | |
| "loss": 0.3115, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 2.8279952847845123e-06, | |
| "loss": 0.3143, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 2.7785755433336436e-06, | |
| "loss": 0.3086, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.7291558018827745e-06, | |
| "loss": 0.3295, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.679736060431906e-06, | |
| "loss": 0.3056, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.8321005917159763, | |
| "eval_loss": 0.32298120856285095, | |
| "eval_runtime": 8.1227, | |
| "eval_samples_per_second": 487.029, | |
| "eval_steps_per_second": 7.633, | |
| "step": 8903 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.630316318981037e-06, | |
| "loss": 0.318, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 2.580896577530169e-06, | |
| "loss": 0.2982, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 2.5314768360792997e-06, | |
| "loss": 0.2874, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.482057094628431e-06, | |
| "loss": 0.2856, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 2.4326373531775623e-06, | |
| "loss": 0.3088, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 2.3832176117266936e-06, | |
| "loss": 0.2887, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.333797870275825e-06, | |
| "loss": 0.2844, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.2843781288249563e-06, | |
| "loss": 0.2884, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 2.2349583873740876e-06, | |
| "loss": 0.2838, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 2.1855386459232185e-06, | |
| "loss": 0.2988, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 2.1361189044723498e-06, | |
| "loss": 0.2881, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 2.0866991630214815e-06, | |
| "loss": 0.2822, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 2.0372794215706124e-06, | |
| "loss": 0.2904, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 1.9878596801197437e-06, | |
| "loss": 0.289, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.938439938668875e-06, | |
| "loss": 0.3042, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.889020197218006e-06, | |
| "loss": 0.286, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 1.8396004557671378e-06, | |
| "loss": 0.2667, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 1.790180714316269e-06, | |
| "loss": 0.3005, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.8438514244500541, | |
| "eval_loss": 0.33582502603530884, | |
| "eval_runtime": 8.1185, | |
| "eval_samples_per_second": 487.283, | |
| "eval_steps_per_second": 7.637, | |
| "step": 17806 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.7407609728654002e-06, | |
| "loss": 0.2686, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.6913412314145313e-06, | |
| "loss": 0.2623, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.6419214899636626e-06, | |
| "loss": 0.2821, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.592501748512794e-06, | |
| "loss": 0.2601, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.5430820070619253e-06, | |
| "loss": 0.2689, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.4936622656110566e-06, | |
| "loss": 0.2583, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 1.4442425241601877e-06, | |
| "loss": 0.2699, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 1.3948227827093192e-06, | |
| "loss": 0.2702, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.3454030412584503e-06, | |
| "loss": 0.2668, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 1.2959832998075814e-06, | |
| "loss": 0.2564, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 1.2465635583567129e-06, | |
| "loss": 0.2677, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 1.197143816905844e-06, | |
| "loss": 0.2478, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 1.1477240754549755e-06, | |
| "loss": 0.268, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.0983043340041066e-06, | |
| "loss": 0.2648, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.048884592553238e-06, | |
| "loss": 0.2334, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 9.994648511023692e-07, | |
| "loss": 0.2636, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 9.500451096515004e-07, | |
| "loss": 0.276, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 9.006253682006318e-07, | |
| "loss": 0.2459, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.8519724936663048, | |
| "eval_loss": 0.351912260055542, | |
| "eval_runtime": 8.1263, | |
| "eval_samples_per_second": 486.814, | |
| "eval_steps_per_second": 7.63, | |
| "step": 26709 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.51205626749763e-07, | |
| "loss": 0.253, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 8.017858852988942e-07, | |
| "loss": 0.2467, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 7.523661438480254e-07, | |
| "loss": 0.2241, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 7.029464023971568e-07, | |
| "loss": 0.2455, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 6.535266609462881e-07, | |
| "loss": 0.2601, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 6.041069194954194e-07, | |
| "loss": 0.2436, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 5.546871780445507e-07, | |
| "loss": 0.2563, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 5.052674365936819e-07, | |
| "loss": 0.236, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 4.5584769514281314e-07, | |
| "loss": 0.2366, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 4.0642795369194445e-07, | |
| "loss": 0.2328, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 3.570082122410757e-07, | |
| "loss": 0.2576, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 3.0758847079020696e-07, | |
| "loss": 0.2562, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 2.581687293393382e-07, | |
| "loss": 0.2305, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 2.087489878884695e-07, | |
| "loss": 0.2405, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 1.5932924643760078e-07, | |
| "loss": 0.2642, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 1.0990950498673206e-07, | |
| "loss": 0.2582, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 6.048976353586333e-08, | |
| "loss": 0.2307, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 1.1070022084994596e-08, | |
| "loss": 0.2361, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.8551470588235294, | |
| "eval_loss": 0.3704700767993927, | |
| "eval_runtime": 8.3254, | |
| "eval_samples_per_second": 475.172, | |
| "eval_steps_per_second": 7.447, | |
| "step": 35612 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 35612, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 1.5744961941257604e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 3.5198716650966744e-06, | |
| "num_train_epochs": 4, | |
| "per_device_train_batch_size": 8, | |
| "seed": 35 | |
| } | |
| } | |