| { | |
| "best_metric": 7.84664249420166, | |
| "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/fict-full-lstm-42/checkpoints/checkpoint-120", | |
| "epoch": 0.96, | |
| "eval_steps": 10, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 10.8142, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 10.805203437805176, | |
| "eval_runtime": 2.7628, | |
| "eval_samples_per_second": 361.95, | |
| "eval_steps_per_second": 45.244, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.2e-05, | |
| "loss": 10.7959, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 10.783019065856934, | |
| "eval_runtime": 2.9141, | |
| "eval_samples_per_second": 343.159, | |
| "eval_steps_per_second": 42.895, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 3.8e-05, | |
| "loss": 10.7637, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 10.725115776062012, | |
| "eval_runtime": 2.7934, | |
| "eval_samples_per_second": 357.988, | |
| "eval_steps_per_second": 44.748, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 10.62, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 10.306303977966309, | |
| "eval_runtime": 2.7909, | |
| "eval_samples_per_second": 358.31, | |
| "eval_steps_per_second": 44.789, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3e-05, | |
| "loss": 9.8971, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 9.273224830627441, | |
| "eval_runtime": 2.8125, | |
| "eval_samples_per_second": 355.55, | |
| "eval_steps_per_second": 44.444, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 9.0782, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 8.667804718017578, | |
| "eval_runtime": 2.8983, | |
| "eval_samples_per_second": 345.031, | |
| "eval_steps_per_second": 43.129, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 8.5836, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 8.331869125366211, | |
| "eval_runtime": 2.7654, | |
| "eval_samples_per_second": 361.611, | |
| "eval_steps_per_second": 45.201, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.8e-05, | |
| "loss": 8.4033, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 8.129607200622559, | |
| "eval_runtime": 2.7665, | |
| "eval_samples_per_second": 361.474, | |
| "eval_steps_per_second": 45.184, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 8.2454, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 8.00214672088623, | |
| "eval_runtime": 2.7731, | |
| "eval_samples_per_second": 360.605, | |
| "eval_steps_per_second": 45.076, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1e-05, | |
| "loss": 8.1265, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 7.9211931228637695, | |
| "eval_runtime": 2.7627, | |
| "eval_samples_per_second": 361.96, | |
| "eval_steps_per_second": 45.245, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 6e-06, | |
| "loss": 8.0655, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 7.871333122253418, | |
| "eval_runtime": 2.7896, | |
| "eval_samples_per_second": 358.475, | |
| "eval_steps_per_second": 44.809, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 8.1315, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 7.84664249420166, | |
| "eval_runtime": 2.9029, | |
| "eval_samples_per_second": 344.487, | |
| "eval_steps_per_second": 43.061, | |
| "step": 120 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 125, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "total_flos": 8440665834240.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |