| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.21691973969631237, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004338394793926247, | |
| "eval_loss": 4.225423812866211, | |
| "eval_runtime": 3.4128, | |
| "eval_samples_per_second": 113.691, | |
| "eval_steps_per_second": 14.358, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.021691973969631236, | |
| "grad_norm": 3.2278621196746826, | |
| "learning_rate": 5e-05, | |
| "loss": 3.9646, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04338394793926247, | |
| "grad_norm": 2.336613416671753, | |
| "learning_rate": 0.0001, | |
| "loss": 3.902, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04338394793926247, | |
| "eval_loss": 3.702977180480957, | |
| "eval_runtime": 1.9932, | |
| "eval_samples_per_second": 194.659, | |
| "eval_steps_per_second": 24.583, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0650759219088937, | |
| "grad_norm": 2.4833273887634277, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.4104, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08676789587852494, | |
| "grad_norm": 3.063148260116577, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.9585, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08676789587852494, | |
| "eval_loss": 2.912097930908203, | |
| "eval_runtime": 1.9963, | |
| "eval_samples_per_second": 194.364, | |
| "eval_steps_per_second": 24.546, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10845986984815618, | |
| "grad_norm": 2.657841205596924, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.7878, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1301518438177874, | |
| "grad_norm": 2.098438024520874, | |
| "learning_rate": 5e-05, | |
| "loss": 2.4937, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1301518438177874, | |
| "eval_loss": 2.6806929111480713, | |
| "eval_runtime": 2.0353, | |
| "eval_samples_per_second": 190.638, | |
| "eval_steps_per_second": 24.075, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15184381778741865, | |
| "grad_norm": 2.968818426132202, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 2.4681, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.1735357917570499, | |
| "grad_norm": 2.618274450302124, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 2.5978, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1735357917570499, | |
| "eval_loss": 2.615467071533203, | |
| "eval_runtime": 2.0059, | |
| "eval_samples_per_second": 193.427, | |
| "eval_steps_per_second": 24.428, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.19522776572668113, | |
| "grad_norm": 2.8834228515625, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 2.5733, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.21691973969631237, | |
| "grad_norm": 2.6873843669891357, | |
| "learning_rate": 0.0, | |
| "loss": 2.5299, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.21691973969631237, | |
| "eval_loss": 2.605391502380371, | |
| "eval_runtime": 1.9911, | |
| "eval_samples_per_second": 194.864, | |
| "eval_steps_per_second": 24.609, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2466357111160832.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |