| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.217932224273682, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5643, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7042606516290727, | |
| "eval_f1": 0.5981941694544958, | |
| "eval_loss": 0.5218127369880676, | |
| "eval_precision": 0.6240824392998306, | |
| "eval_recall": 0.5932442262229496, | |
| "eval_runtime": 1.7249, | |
| "eval_samples_per_second": 231.323, | |
| "eval_steps_per_second": 28.988, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.5043983459472656, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5086, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_f1": 0.6868131868131868, | |
| "eval_loss": 0.5021316409111023, | |
| "eval_precision": 0.6814393939393939, | |
| "eval_recall": 0.6959901800327333, | |
| "eval_runtime": 1.7316, | |
| "eval_samples_per_second": 230.424, | |
| "eval_steps_per_second": 28.875, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.2947137355804443, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4652, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7894736842105263, | |
| "eval_f1": 0.7407327186088248, | |
| "eval_loss": 0.444990336894989, | |
| "eval_precision": 0.7463777285669092, | |
| "eval_recall": 0.7360429168939807, | |
| "eval_runtime": 1.7474, | |
| "eval_samples_per_second": 228.342, | |
| "eval_steps_per_second": 28.614, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.659485816955566, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4248, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.7917273014868713, | |
| "eval_loss": 0.39323920011520386, | |
| "eval_precision": 0.8074456774536514, | |
| "eval_recall": 0.780460083651573, | |
| "eval_runtime": 1.7487, | |
| "eval_samples_per_second": 228.174, | |
| "eval_steps_per_second": 28.593, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.222157955169678, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3812, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8118502107020158, | |
| "eval_loss": 0.37043145298957825, | |
| "eval_precision": 0.8083091673078061, | |
| "eval_recall": 0.8157846881251136, | |
| "eval_runtime": 1.756, | |
| "eval_samples_per_second": 227.224, | |
| "eval_steps_per_second": 28.474, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 6.741718769073486, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3506, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8289446964056049, | |
| "eval_loss": 0.35664382576942444, | |
| "eval_precision": 0.8266129032258065, | |
| "eval_recall": 0.8314238952536825, | |
| "eval_runtime": 1.751, | |
| "eval_samples_per_second": 227.875, | |
| "eval_steps_per_second": 28.556, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.737555503845215, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3323, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.820640215771169, | |
| "eval_loss": 0.3437691330909729, | |
| "eval_precision": 0.8364527629233511, | |
| "eval_recall": 0.8089198036006546, | |
| "eval_runtime": 1.756, | |
| "eval_samples_per_second": 227.216, | |
| "eval_steps_per_second": 28.473, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 6.206509113311768, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3108, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8279052989013229, | |
| "eval_loss": 0.33262690901756287, | |
| "eval_precision": 0.8414113428943938, | |
| "eval_recall": 0.8174668121476631, | |
| "eval_runtime": 1.7517, | |
| "eval_samples_per_second": 227.78, | |
| "eval_steps_per_second": 28.544, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.5029311180114746, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2998, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8385441718775052, | |
| "eval_loss": 0.32504233717918396, | |
| "eval_precision": 0.8412280701754387, | |
| "eval_recall": 0.8360156392071285, | |
| "eval_runtime": 1.7542, | |
| "eval_samples_per_second": 227.46, | |
| "eval_steps_per_second": 28.504, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.729393005371094, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2923, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8263588263588264, | |
| "eval_loss": 0.31815892457962036, | |
| "eval_precision": 0.8289473684210527, | |
| "eval_recall": 0.8239225313693399, | |
| "eval_runtime": 1.7579, | |
| "eval_samples_per_second": 226.98, | |
| "eval_steps_per_second": 28.444, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 11.397634506225586, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2887, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8438308224802573, | |
| "eval_loss": 0.31445056200027466, | |
| "eval_precision": 0.8485409407665505, | |
| "eval_recall": 0.8395617384979087, | |
| "eval_runtime": 1.7554, | |
| "eval_samples_per_second": 227.301, | |
| "eval_steps_per_second": 28.484, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 10.077719688415527, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2716, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8430067043674827, | |
| "eval_loss": 0.30924713611602783, | |
| "eval_precision": 0.8498269896193772, | |
| "eval_recall": 0.8370612838697945, | |
| "eval_runtime": 1.7521, | |
| "eval_samples_per_second": 227.724, | |
| "eval_steps_per_second": 28.537, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 3.046454906463623, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2598, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8458135188208294, | |
| "eval_loss": 0.30985820293426514, | |
| "eval_precision": 0.8627946127946129, | |
| "eval_recall": 0.8331060192762321, | |
| "eval_runtime": 1.754, | |
| "eval_samples_per_second": 227.479, | |
| "eval_steps_per_second": 28.506, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 7.8712358474731445, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2722, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8491633041962088, | |
| "eval_loss": 0.3003251254558563, | |
| "eval_precision": 0.85610254797106, | |
| "eval_recall": 0.8431078377886888, | |
| "eval_runtime": 1.7541, | |
| "eval_samples_per_second": 227.468, | |
| "eval_steps_per_second": 28.505, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.7105746269226074, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2536, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8491633041962088, | |
| "eval_loss": 0.29776889085769653, | |
| "eval_precision": 0.85610254797106, | |
| "eval_recall": 0.8431078377886888, | |
| "eval_runtime": 1.7521, | |
| "eval_samples_per_second": 227.726, | |
| "eval_steps_per_second": 28.537, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 3.3939132690429688, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2536, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8568221901555235, | |
| "eval_loss": 0.2970341145992279, | |
| "eval_precision": 0.8596491228070176, | |
| "eval_recall": 0.8541553009638116, | |
| "eval_runtime": 1.757, | |
| "eval_samples_per_second": 227.091, | |
| "eval_steps_per_second": 28.457, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.2166881561279297, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2479, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8545433391506689, | |
| "eval_loss": 0.29783862829208374, | |
| "eval_precision": 0.8639270714012982, | |
| "eval_recall": 0.846653937079469, | |
| "eval_runtime": 1.7549, | |
| "eval_samples_per_second": 227.365, | |
| "eval_steps_per_second": 28.492, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 7.031806945800781, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2487, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8491633041962088, | |
| "eval_loss": 0.29700446128845215, | |
| "eval_precision": 0.85610254797106, | |
| "eval_recall": 0.8431078377886888, | |
| "eval_runtime": 1.7527, | |
| "eval_samples_per_second": 227.643, | |
| "eval_steps_per_second": 28.527, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 5.178986549377441, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2457, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8421640488656195, | |
| "eval_loss": 0.2947324812412262, | |
| "eval_precision": 0.8512313860252005, | |
| "eval_recall": 0.8345608292416803, | |
| "eval_runtime": 1.7547, | |
| "eval_samples_per_second": 227.393, | |
| "eval_steps_per_second": 28.495, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.935147285461426, | |
| "learning_rate": 0.0, | |
| "loss": 0.2499, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8421640488656195, | |
| "eval_loss": 0.29513633251190186, | |
| "eval_precision": 0.8512313860252005, | |
| "eval_recall": 0.8345608292416803, | |
| "eval_runtime": 1.7593, | |
| "eval_samples_per_second": 226.794, | |
| "eval_steps_per_second": 28.42, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7609911792720000.0, | |
| "train_loss": 0.32608639607664014, | |
| "train_runtime": 629.3107, | |
| "train_samples_per_second": 115.619, | |
| "train_steps_per_second": 3.877 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7609911792720000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |