| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06054872280037843, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0030274361400189215, | |
| "grad_norm": 5.019101619720459, | |
| "learning_rate": 1e-05, | |
| "loss": 7.4266, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0030274361400189215, | |
| "eval_loss": 7.595103740692139, | |
| "eval_runtime": 54.9308, | |
| "eval_samples_per_second": 40.524, | |
| "eval_steps_per_second": 10.14, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006054872280037843, | |
| "grad_norm": 4.992447853088379, | |
| "learning_rate": 2e-05, | |
| "loss": 7.5358, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.009082308420056764, | |
| "grad_norm": 5.094038963317871, | |
| "learning_rate": 3e-05, | |
| "loss": 7.4946, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.012109744560075686, | |
| "grad_norm": 4.877178192138672, | |
| "learning_rate": 4e-05, | |
| "loss": 7.4822, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.015137180700094607, | |
| "grad_norm": 5.0338568687438965, | |
| "learning_rate": 5e-05, | |
| "loss": 7.3274, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015137180700094607, | |
| "eval_loss": 7.441568374633789, | |
| "eval_runtime": 54.9164, | |
| "eval_samples_per_second": 40.534, | |
| "eval_steps_per_second": 10.143, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01816461684011353, | |
| "grad_norm": 5.345804691314697, | |
| "learning_rate": 6e-05, | |
| "loss": 7.3461, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.02119205298013245, | |
| "grad_norm": 5.390858173370361, | |
| "learning_rate": 7e-05, | |
| "loss": 7.2021, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.024219489120151372, | |
| "grad_norm": 5.532158374786377, | |
| "learning_rate": 8e-05, | |
| "loss": 7.0189, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.027246925260170293, | |
| "grad_norm": 5.653778076171875, | |
| "learning_rate": 9e-05, | |
| "loss": 6.9335, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.030274361400189215, | |
| "grad_norm": 5.158017158508301, | |
| "learning_rate": 0.0001, | |
| "loss": 6.6725, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.030274361400189215, | |
| "eval_loss": 6.509109020233154, | |
| "eval_runtime": 54.932, | |
| "eval_samples_per_second": 40.523, | |
| "eval_steps_per_second": 10.14, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.033301797540208136, | |
| "grad_norm": 4.311172962188721, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 6.464, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03632923368022706, | |
| "grad_norm": 4.248252868652344, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 6.1797, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03935666982024598, | |
| "grad_norm": 4.477457046508789, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 6.0317, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0423841059602649, | |
| "grad_norm": 3.6723344326019287, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 5.8527, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.04541154210028382, | |
| "grad_norm": 3.421534538269043, | |
| "learning_rate": 5e-05, | |
| "loss": 5.7074, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04541154210028382, | |
| "eval_loss": 5.635517597198486, | |
| "eval_runtime": 54.9282, | |
| "eval_samples_per_second": 40.526, | |
| "eval_steps_per_second": 10.141, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.048438978240302744, | |
| "grad_norm": 2.373675584793091, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 5.585, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.051466414380321665, | |
| "grad_norm": 2.601724624633789, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 5.5902, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.054493850520340587, | |
| "grad_norm": 2.4370365142822266, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 5.51, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05752128666035951, | |
| "grad_norm": 2.936497449874878, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 5.4948, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.06054872280037843, | |
| "grad_norm": 3.4013614654541016, | |
| "learning_rate": 0.0, | |
| "loss": 5.4982, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06054872280037843, | |
| "eval_loss": 5.492336750030518, | |
| "eval_runtime": 54.9414, | |
| "eval_samples_per_second": 40.516, | |
| "eval_steps_per_second": 10.138, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "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": 6.587960141021184e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |