| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 2.11942007802131, | |
| "learning_rate": 4e-05, | |
| "loss": 2.9577, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.5155914071960075, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6444, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 2.1252453327178955, | |
| "eval_runtime": 3.4068, | |
| "eval_samples_per_second": 5.871, | |
| "eval_steps_per_second": 1.468, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.572036026319978, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 1.6642, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.4520132541656494, | |
| "eval_runtime": 1.9411, | |
| "eval_samples_per_second": 10.303, | |
| "eval_steps_per_second": 2.576, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.5888307849172969, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 1.29, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 1.241407871246338, | |
| "eval_runtime": 1.9398, | |
| "eval_samples_per_second": 10.31, | |
| "eval_steps_per_second": 2.578, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.5493807394559196, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.151, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.1817108392715454, | |
| "eval_runtime": 1.9405, | |
| "eval_samples_per_second": 10.307, | |
| "eval_steps_per_second": 2.577, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.5681855042713849, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.0661, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1291143894195557, | |
| "eval_runtime": 1.9401, | |
| "eval_samples_per_second": 10.309, | |
| "eval_steps_per_second": 2.577, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.39762263218255406, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.9829, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.1112431287765503, | |
| "eval_runtime": 1.9347, | |
| "eval_samples_per_second": 10.337, | |
| "eval_steps_per_second": 2.584, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.4041256873727563, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.9062, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 1.1073366403579712, | |
| "eval_runtime": 1.9418, | |
| "eval_samples_per_second": 10.3, | |
| "eval_steps_per_second": 2.575, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.4112568137648661, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.8685, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 1.1047579050064087, | |
| "eval_runtime": 1.9408, | |
| "eval_samples_per_second": 10.305, | |
| "eval_steps_per_second": 2.576, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.38922449035685425, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 0.8852, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "eval_loss": 1.1037671566009521, | |
| "eval_runtime": 1.9392, | |
| "eval_samples_per_second": 10.313, | |
| "eval_steps_per_second": 2.578, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.41298541442927883, | |
| "learning_rate": 0.0, | |
| "loss": 0.7973, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.103836178779602, | |
| "eval_runtime": 1.9419, | |
| "eval_samples_per_second": 10.299, | |
| "eval_steps_per_second": 2.575, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 50, | |
| "total_flos": 261193039872.0, | |
| "train_loss": 1.231847505569458, | |
| "train_runtime": 175.0963, | |
| "train_samples_per_second": 2.284, | |
| "train_steps_per_second": 0.286 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 261193039872.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |