| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9333333333333336, | |
| "eval_steps": 5, | |
| "global_step": 33, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.9699, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.9491, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7319, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.5441049337387085, | |
| "eval_runtime": 8.6083, | |
| "eval_samples_per_second": 18.122, | |
| "eval_steps_per_second": 0.581, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.941551389275217e-05, | |
| "loss": 1.5341, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.768938549177393e-05, | |
| "loss": 1.4557, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.4902326642641095e-05, | |
| "loss": 1.3425, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 1.2967758178710938, | |
| "eval_runtime": 8.8757, | |
| "eval_samples_per_second": 17.576, | |
| "eval_steps_per_second": 0.563, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.118465711954569e-05, | |
| "loss": 1.2552, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.671021101749476e-05, | |
| "loss": 1.1709, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 1.2151137590408325, | |
| "eval_runtime": 8.1047, | |
| "eval_samples_per_second": 19.248, | |
| "eval_steps_per_second": 0.617, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 3.168820846323053e-05, | |
| "loss": 1.1595, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.635347271463544e-05, | |
| "loss": 1.1061, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 2.0955450086180882e-05, | |
| "loss": 1.0994, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_loss": 1.160514235496521, | |
| "eval_runtime": 7.964, | |
| "eval_samples_per_second": 19.588, | |
| "eval_steps_per_second": 0.628, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 1.574654611650214e-05, | |
| "loss": 1.0804, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.0970323365940444e-05, | |
| "loss": 1.0287, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_loss": 1.1381587982177734, | |
| "eval_runtime": 7.9285, | |
| "eval_samples_per_second": 19.676, | |
| "eval_steps_per_second": 0.631, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 6.8501127019217346e-06, | |
| "loss": 1.0266, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 3.578570595810274e-06, | |
| "loss": 1.0038, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 1.3086707204299414e-06, | |
| "loss": 1.0303, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 1.1314786672592163, | |
| "eval_runtime": 7.9656, | |
| "eval_samples_per_second": 19.584, | |
| "eval_steps_per_second": 0.628, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.4655107114101007e-07, | |
| "loss": 1.0081, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "step": 33, | |
| "total_flos": 17934172815360.0, | |
| "train_loss": 1.2418649593989055, | |
| "train_runtime": 753.4405, | |
| "train_samples_per_second": 5.626, | |
| "train_steps_per_second": 0.044 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 33, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 17934172815360.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |