| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 12.0, | |
| "global_step": 6144, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 4.755859375e-05, | |
| "loss": 228.0322, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 95.51275634765625, | |
| "eval_runtime": 19.2239, | |
| "eval_samples_per_second": 84.738, | |
| "eval_steps_per_second": 5.306, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.5117187500000005e-05, | |
| "loss": 102.3494, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 94.07129669189453, | |
| "eval_runtime": 19.254, | |
| "eval_samples_per_second": 84.606, | |
| "eval_steps_per_second": 5.298, | |
| "step": 1024 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.267578125e-05, | |
| "loss": 90.6642, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 83.28936767578125, | |
| "eval_runtime": 19.2397, | |
| "eval_samples_per_second": 84.669, | |
| "eval_steps_per_second": 5.302, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.0234375e-05, | |
| "loss": 78.5213, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 83.8617172241211, | |
| "eval_runtime": 19.2535, | |
| "eval_samples_per_second": 84.608, | |
| "eval_steps_per_second": 5.298, | |
| "step": 2048 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 3.7792968750000005e-05, | |
| "loss": 70.4173, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 79.97139739990234, | |
| "eval_runtime": 19.2503, | |
| "eval_samples_per_second": 84.622, | |
| "eval_steps_per_second": 5.299, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 3.53515625e-05, | |
| "loss": 63.1215, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 81.74008178710938, | |
| "eval_runtime": 19.2347, | |
| "eval_samples_per_second": 84.69, | |
| "eval_steps_per_second": 5.303, | |
| "step": 3072 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 3.291015625e-05, | |
| "loss": 55.0817, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 80.47997283935547, | |
| "eval_runtime": 19.239, | |
| "eval_samples_per_second": 84.672, | |
| "eval_steps_per_second": 5.302, | |
| "step": 3584 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 3.0468750000000002e-05, | |
| "loss": 49.4482, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 78.9246597290039, | |
| "eval_runtime": 19.2369, | |
| "eval_samples_per_second": 84.681, | |
| "eval_steps_per_second": 5.302, | |
| "step": 4096 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "learning_rate": 2.802734375e-05, | |
| "loss": 45.7534, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 79.2200927734375, | |
| "eval_runtime": 19.2428, | |
| "eval_samples_per_second": 84.655, | |
| "eval_steps_per_second": 5.301, | |
| "step": 4608 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 2.55859375e-05, | |
| "loss": 43.1915, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 81.24749755859375, | |
| "eval_runtime": 19.2441, | |
| "eval_samples_per_second": 84.649, | |
| "eval_steps_per_second": 5.3, | |
| "step": 5120 | |
| }, | |
| { | |
| "epoch": 10.74, | |
| "learning_rate": 2.3144531250000002e-05, | |
| "loss": 41.0389, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 78.5303955078125, | |
| "eval_runtime": 19.2504, | |
| "eval_samples_per_second": 84.622, | |
| "eval_steps_per_second": 5.299, | |
| "step": 5632 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 2.0703125e-05, | |
| "loss": 38.4756, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 77.77259826660156, | |
| "eval_runtime": 19.2175, | |
| "eval_samples_per_second": 84.766, | |
| "eval_steps_per_second": 5.308, | |
| "step": 6144 | |
| } | |
| ], | |
| "max_steps": 10240, | |
| "num_train_epochs": 20, | |
| "total_flos": 2.582451944655667e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |