| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.43103448275862066, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008620689655172414, | |
| "eval_loss": 1.9068092107772827, | |
| "eval_runtime": 17.6833, | |
| "eval_samples_per_second": 11.027, | |
| "eval_steps_per_second": 1.414, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02586206896551724, | |
| "grad_norm": 42.144020080566406, | |
| "learning_rate": 3e-05, | |
| "loss": 7.5638, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04310344827586207, | |
| "eval_loss": 1.08273184299469, | |
| "eval_runtime": 17.8695, | |
| "eval_samples_per_second": 10.912, | |
| "eval_steps_per_second": 1.399, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05172413793103448, | |
| "grad_norm": 23.90378761291504, | |
| "learning_rate": 6e-05, | |
| "loss": 5.3623, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07758620689655173, | |
| "grad_norm": 28.774568557739258, | |
| "learning_rate": 9e-05, | |
| "loss": 4.0848, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08620689655172414, | |
| "eval_loss": 0.7605625987052917, | |
| "eval_runtime": 17.9379, | |
| "eval_samples_per_second": 10.871, | |
| "eval_steps_per_second": 1.394, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10344827586206896, | |
| "grad_norm": 12.58159065246582, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 3.2548, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12931034482758622, | |
| "grad_norm": 14.037184715270996, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.3872, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.12931034482758622, | |
| "eval_loss": 0.710814893245697, | |
| "eval_runtime": 17.9839, | |
| "eval_samples_per_second": 10.843, | |
| "eval_steps_per_second": 1.39, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.15517241379310345, | |
| "grad_norm": 12.68582534790039, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 3.2727, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1724137931034483, | |
| "eval_loss": 0.6887093782424927, | |
| "eval_runtime": 17.9522, | |
| "eval_samples_per_second": 10.862, | |
| "eval_steps_per_second": 1.393, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1810344827586207, | |
| "grad_norm": 12.186010360717773, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 2.8286, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.20689655172413793, | |
| "grad_norm": 14.021751403808594, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 2.9939, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.21551724137931033, | |
| "eval_loss": 0.6674009561538696, | |
| "eval_runtime": 17.963, | |
| "eval_samples_per_second": 10.856, | |
| "eval_steps_per_second": 1.392, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.23275862068965517, | |
| "grad_norm": 9.224210739135742, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 2.5492, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.25862068965517243, | |
| "grad_norm": 10.450089454650879, | |
| "learning_rate": 5e-05, | |
| "loss": 2.644, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25862068965517243, | |
| "eval_loss": 0.6499032974243164, | |
| "eval_runtime": 17.9607, | |
| "eval_samples_per_second": 10.857, | |
| "eval_steps_per_second": 1.392, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28448275862068967, | |
| "grad_norm": 11.339786529541016, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 2.8155, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3017241379310345, | |
| "eval_loss": 0.648213803768158, | |
| "eval_runtime": 17.968, | |
| "eval_samples_per_second": 10.853, | |
| "eval_steps_per_second": 1.391, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3103448275862069, | |
| "grad_norm": 11.725385665893555, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 2.455, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.33620689655172414, | |
| "grad_norm": 12.540617942810059, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 3.135, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.3448275862068966, | |
| "eval_loss": 0.6416993141174316, | |
| "eval_runtime": 17.9633, | |
| "eval_samples_per_second": 10.855, | |
| "eval_steps_per_second": 1.392, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3620689655172414, | |
| "grad_norm": 12.699934959411621, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.5006, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.3879310344827586, | |
| "grad_norm": 12.192987442016602, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 2.9317, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3879310344827586, | |
| "eval_loss": 0.6367151737213135, | |
| "eval_runtime": 17.9539, | |
| "eval_samples_per_second": 10.861, | |
| "eval_steps_per_second": 1.392, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.41379310344827586, | |
| "grad_norm": 11.72185230255127, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 2.7245, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.43103448275862066, | |
| "eval_loss": 0.6354005336761475, | |
| "eval_runtime": 17.9512, | |
| "eval_samples_per_second": 10.863, | |
| "eval_steps_per_second": 1.393, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.15499004493824e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |