| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7518796992481203, | |
| "eval_steps": 63, | |
| "global_step": 250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0030075187969924814, | |
| "eval_loss": 10.374281883239746, | |
| "eval_runtime": 0.3586, | |
| "eval_samples_per_second": 390.36, | |
| "eval_steps_per_second": 195.18, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03007518796992481, | |
| "grad_norm": 0.060663752257823944, | |
| "learning_rate": 0.00019979453927503364, | |
| "loss": 10.3728, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06015037593984962, | |
| "grad_norm": 0.05574655905365944, | |
| "learning_rate": 0.00019815591569910654, | |
| "loss": 10.3704, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09022556390977443, | |
| "grad_norm": 0.07972452044487, | |
| "learning_rate": 0.00019490557470106686, | |
| "loss": 10.3695, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12030075187969924, | |
| "grad_norm": 0.14503557980060577, | |
| "learning_rate": 0.0001900968867902419, | |
| "loss": 10.3621, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15037593984962405, | |
| "grad_norm": 0.13525503873825073, | |
| "learning_rate": 0.00018380881048918405, | |
| "loss": 10.3626, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18045112781954886, | |
| "grad_norm": 0.09932214021682739, | |
| "learning_rate": 0.00017614459583691346, | |
| "loss": 10.3584, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18947368421052632, | |
| "eval_loss": 10.356450080871582, | |
| "eval_runtime": 0.3534, | |
| "eval_samples_per_second": 396.183, | |
| "eval_steps_per_second": 198.091, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 0.1384136825799942, | |
| "learning_rate": 0.0001672300890261317, | |
| "loss": 10.3572, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24060150375939848, | |
| "grad_norm": 0.1283988505601883, | |
| "learning_rate": 0.00015721166601221698, | |
| "loss": 10.353, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2706766917293233, | |
| "grad_norm": 0.11646091192960739, | |
| "learning_rate": 0.00014625382902408356, | |
| "loss": 10.351, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3007518796992481, | |
| "grad_norm": 0.1207568570971489, | |
| "learning_rate": 0.00013453650544213076, | |
| "loss": 10.3498, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3308270676691729, | |
| "grad_norm": 0.09924621134996414, | |
| "learning_rate": 0.00012225209339563145, | |
| "loss": 10.3497, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3609022556390977, | |
| "grad_norm": 0.08627820760011673, | |
| "learning_rate": 0.00010960230259076818, | |
| "loss": 10.3494, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.37894736842105264, | |
| "eval_loss": 10.348790168762207, | |
| "eval_runtime": 0.3677, | |
| "eval_samples_per_second": 380.716, | |
| "eval_steps_per_second": 190.358, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.39097744360902253, | |
| "grad_norm": 0.1328861266374588, | |
| "learning_rate": 9.679484224283449e-05, | |
| "loss": 10.3492, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "grad_norm": 0.0968804806470871, | |
| "learning_rate": 8.404001049666211e-05, | |
| "loss": 10.3483, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.45112781954887216, | |
| "grad_norm": 0.11394359171390533, | |
| "learning_rate": 7.154724133689677e-05, | |
| "loss": 10.347, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.48120300751879697, | |
| "grad_norm": 0.1099514290690422, | |
| "learning_rate": 5.952166568776062e-05, | |
| "loss": 10.3473, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5112781954887218, | |
| "grad_norm": 0.08238931745290756, | |
| "learning_rate": 4.8160743168947496e-05, | |
| "loss": 10.3464, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5413533834586466, | |
| "grad_norm": 0.10015811026096344, | |
| "learning_rate": 3.7651019814126654e-05, | |
| "loss": 10.3481, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5684210526315789, | |
| "eval_loss": 10.34677791595459, | |
| "eval_runtime": 0.365, | |
| "eval_samples_per_second": 383.611, | |
| "eval_steps_per_second": 191.806, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.10194003582000732, | |
| "learning_rate": 2.8165064990227252e-05, | |
| "loss": 10.3449, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6015037593984962, | |
| "grad_norm": 0.12122286856174469, | |
| "learning_rate": 1.985863781320435e-05, | |
| "loss": 10.3472, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 0.07524222135543823, | |
| "learning_rate": 1.286812958766106e-05, | |
| "loss": 10.3454, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6616541353383458, | |
| "grad_norm": 0.12044119089841843, | |
| "learning_rate": 7.308324265397836e-06, | |
| "loss": 10.3452, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6917293233082706, | |
| "grad_norm": 0.1054227277636528, | |
| "learning_rate": 3.270513696097055e-06, | |
| "loss": 10.3468, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7218045112781954, | |
| "grad_norm": 0.07029376924037933, | |
| "learning_rate": 8.209986176753948e-07, | |
| "loss": 10.3477, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7518796992481203, | |
| "grad_norm": 0.07956445217132568, | |
| "learning_rate": 0.0, | |
| "loss": 10.3464, | |
| "step": 250 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 63, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3232727040000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |