| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9797979797979797, | |
| "eval_steps": 500, | |
| "global_step": 147, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04040404040404041, | |
| "grad_norm": 2.354057369065775, | |
| "learning_rate": 2.5e-05, | |
| "loss": 3.6821, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.20202020202020202, | |
| "grad_norm": 3.318817777883566, | |
| "learning_rate": 0.000125, | |
| "loss": 3.6411, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.40404040404040403, | |
| "grad_norm": 1.1567295340838406, | |
| "learning_rate": 0.0001995184726672197, | |
| "loss": 2.3671, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6060606060606061, | |
| "grad_norm": 0.5651519215666836, | |
| "learning_rate": 0.00019415440651830208, | |
| "loss": 1.478, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8080808080808081, | |
| "grad_norm": 0.4028700304940561, | |
| "learning_rate": 0.00018314696123025454, | |
| "loss": 1.3252, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "eval_loss": 1.1692708730697632, | |
| "eval_runtime": 177.3352, | |
| "eval_samples_per_second": 35.644, | |
| "eval_steps_per_second": 2.233, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0101010101010102, | |
| "grad_norm": 0.3079834505872443, | |
| "learning_rate": 0.00016715589548470185, | |
| "loss": 1.2706, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.2121212121212122, | |
| "grad_norm": 0.25648814038671125, | |
| "learning_rate": 0.0001471396736825998, | |
| "loss": 1.1881, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4141414141414141, | |
| "grad_norm": 0.2008341133631863, | |
| "learning_rate": 0.0001242980179903264, | |
| "loss": 1.1514, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.6161616161616161, | |
| "grad_norm": 0.1814019139468598, | |
| "learning_rate": 0.0001, | |
| "loss": 1.1815, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.18331867012606626, | |
| "learning_rate": 7.570198200967362e-05, | |
| "loss": 1.1352, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.9797979797979797, | |
| "eval_loss": 1.0709240436553955, | |
| "eval_runtime": 175.2955, | |
| "eval_samples_per_second": 36.059, | |
| "eval_steps_per_second": 2.259, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.0202020202020203, | |
| "grad_norm": 0.2143290425477542, | |
| "learning_rate": 5.286032631740023e-05, | |
| "loss": 1.1283, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.18532120253500695, | |
| "learning_rate": 3.2844104515298155e-05, | |
| "loss": 1.112, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.4242424242424243, | |
| "grad_norm": 0.19523394937897012, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 1.1094, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.6262626262626263, | |
| "grad_norm": 0.19014746927385398, | |
| "learning_rate": 5.8455934816979305e-06, | |
| "loss": 1.1021, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.8282828282828283, | |
| "grad_norm": 0.1651175400452345, | |
| "learning_rate": 4.815273327803182e-07, | |
| "loss": 1.1558, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.5252525252525253, | |
| "grad_norm": 1.5131939349650319, | |
| "learning_rate": 0.00011423148382732853, | |
| "loss": 1.4021, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.6262626262626263, | |
| "grad_norm": 0.4578537596053918, | |
| "learning_rate": 0.00010237976975461075, | |
| "loss": 1.2517, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.7272727272727273, | |
| "grad_norm": 0.33225716851066756, | |
| "learning_rate": 9.049439566958175e-05, | |
| "loss": 1.139, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.8282828282828283, | |
| "grad_norm": 0.27743266981716935, | |
| "learning_rate": 7.874347104470234e-05, | |
| "loss": 1.0842, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.9292929292929293, | |
| "grad_norm": 0.3335911253392335, | |
| "learning_rate": 6.729320366825784e-05, | |
| "loss": 1.1265, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.98989898989899, | |
| "eval_loss": 1.0307520627975464, | |
| "eval_runtime": 165.0307, | |
| "eval_samples_per_second": 38.302, | |
| "eval_steps_per_second": 2.4, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.0303030303030303, | |
| "grad_norm": 0.2767977674400862, | |
| "learning_rate": 5.630554876306407e-05, | |
| "loss": 1.0492, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.1313131313131315, | |
| "grad_norm": 0.2747976722762822, | |
| "learning_rate": 4.593591825444028e-05, | |
| "loss": 1.0602, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.2323232323232323, | |
| "grad_norm": 0.2987036396560577, | |
| "learning_rate": 3.6330982588091186e-05, | |
| "loss": 1.0785, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "grad_norm": 0.27812388202445604, | |
| "learning_rate": 2.7626596189492983e-05, | |
| "loss": 1.0563, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.4343434343434343, | |
| "grad_norm": 0.24657888130613284, | |
| "learning_rate": 1.994587590756397e-05, | |
| "loss": 1.0835, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.5353535353535355, | |
| "grad_norm": 0.24337497990605553, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 1.071, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.6363636363636362, | |
| "grad_norm": 0.2979117602370218, | |
| "learning_rate": 8.073969641833445e-06, | |
| "loss": 1.036, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.7373737373737375, | |
| "grad_norm": 0.25304418891814984, | |
| "learning_rate": 4.050702638550275e-06, | |
| "loss": 1.1085, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.8383838383838382, | |
| "grad_norm": 0.26313547369333573, | |
| "learning_rate": 1.3845646281813507e-06, | |
| "loss": 1.0692, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.9393939393939394, | |
| "grad_norm": 0.2581657699360705, | |
| "learning_rate": 1.1326608169920372e-07, | |
| "loss": 1.1113, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.9797979797979797, | |
| "eval_loss": 1.0100449323654175, | |
| "eval_runtime": 162.6209, | |
| "eval_samples_per_second": 38.87, | |
| "eval_steps_per_second": 2.435, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 2.9797979797979797, | |
| "step": 147, | |
| "total_flos": 981865656745984.0, | |
| "train_loss": 0.5657350935903537, | |
| "train_runtime": 914.4964, | |
| "train_samples_per_second": 20.736, | |
| "train_steps_per_second": 0.161 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 147, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 981865656745984.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |