| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.989690721649485, | |
| "eval_steps": 500, | |
| "global_step": 567, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9896907216494846, | |
| "grad_norm": 2.3543875217437744, | |
| "learning_rate": 9.5e-06, | |
| "loss": 3.0813, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.9896907216494846, | |
| "eval_gen_len": 247.969023, | |
| "eval_loss": 2.5624194145202637, | |
| "eval_rouge1": 0.365572, | |
| "eval_rouge2": 0.067814, | |
| "eval_rougeL": 0.174291, | |
| "eval_rougeLsum": 0.338231, | |
| "eval_runtime": 3587.8888, | |
| "eval_samples_per_second": 0.351, | |
| "eval_steps_per_second": 0.088, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.9896907216494846, | |
| "grad_norm": 1.331000566482544, | |
| "learning_rate": 9e-06, | |
| "loss": 2.6117, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.9896907216494846, | |
| "eval_gen_len": 254.757744, | |
| "eval_loss": 2.461825132369995, | |
| "eval_rouge1": 0.378265, | |
| "eval_rouge2": 0.077138, | |
| "eval_rougeL": 0.182503, | |
| "eval_rougeLsum": 0.351427, | |
| "eval_runtime": 3653.2761, | |
| "eval_samples_per_second": 0.345, | |
| "eval_steps_per_second": 0.086, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 2.9896907216494846, | |
| "grad_norm": 0.8951466083526611, | |
| "learning_rate": 8.5e-06, | |
| "loss": 2.5014, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 2.9896907216494846, | |
| "eval_gen_len": 254.992851, | |
| "eval_loss": 2.4188287258148193, | |
| "eval_rouge1": 0.374918, | |
| "eval_rouge2": 0.074194, | |
| "eval_rougeL": 0.179174, | |
| "eval_rougeLsum": 0.347582, | |
| "eval_runtime": 3488.9943, | |
| "eval_samples_per_second": 0.361, | |
| "eval_steps_per_second": 0.09, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 3.9896907216494846, | |
| "grad_norm": 0.8497774004936218, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.4322, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 3.9896907216494846, | |
| "eval_gen_len": 254.067514, | |
| "eval_loss": 2.3958144187927246, | |
| "eval_rouge1": 0.380259, | |
| "eval_rouge2": 0.078095, | |
| "eval_rougeL": 0.183458, | |
| "eval_rougeLsum": 0.353376, | |
| "eval_runtime": 3495.9619, | |
| "eval_samples_per_second": 0.36, | |
| "eval_steps_per_second": 0.09, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 4.989690721649485, | |
| "grad_norm": 0.8451509475708008, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 2.3836, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 4.989690721649485, | |
| "eval_gen_len": 254.285147, | |
| "eval_loss": 2.3810665607452393, | |
| "eval_rouge1": 0.382543, | |
| "eval_rouge2": 0.079173, | |
| "eval_rougeL": 0.183712, | |
| "eval_rougeLsum": 0.355298, | |
| "eval_runtime": 3613.6748, | |
| "eval_samples_per_second": 0.348, | |
| "eval_steps_per_second": 0.087, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 5.989690721649485, | |
| "grad_norm": 0.8150149583816528, | |
| "learning_rate": 7e-06, | |
| "loss": 2.3465, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 5.989690721649485, | |
| "eval_gen_len": 254.706116, | |
| "eval_loss": 2.3701136112213135, | |
| "eval_rouge1": 0.386206, | |
| "eval_rouge2": 0.081617, | |
| "eval_rougeL": 0.186495, | |
| "eval_rougeLsum": 0.359838, | |
| "eval_runtime": 3613.0416, | |
| "eval_samples_per_second": 0.348, | |
| "eval_steps_per_second": 0.087, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 6.989690721649485, | |
| "grad_norm": 0.8424471020698547, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 2.3149, | |
| "step": 441 | |
| }, | |
| { | |
| "epoch": 6.989690721649485, | |
| "eval_gen_len": 254.68467, | |
| "eval_loss": 2.3630220890045166, | |
| "eval_rouge1": 0.388393, | |
| "eval_rouge2": 0.083801, | |
| "eval_rougeL": 0.187039, | |
| "eval_rougeLsum": 0.361264, | |
| "eval_runtime": 3555.8921, | |
| "eval_samples_per_second": 0.354, | |
| "eval_steps_per_second": 0.089, | |
| "step": 441 | |
| }, | |
| { | |
| "epoch": 7.989690721649485, | |
| "grad_norm": 0.8140623569488525, | |
| "learning_rate": 6e-06, | |
| "loss": 2.2876, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 7.989690721649485, | |
| "eval_gen_len": 255.17077, | |
| "eval_loss": 2.356766939163208, | |
| "eval_rouge1": 0.385187, | |
| "eval_rouge2": 0.081047, | |
| "eval_rougeL": 0.185354, | |
| "eval_rougeLsum": 0.357705, | |
| "eval_runtime": 3499.45, | |
| "eval_samples_per_second": 0.36, | |
| "eval_steps_per_second": 0.09, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 8.989690721649485, | |
| "grad_norm": 0.846682071685791, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 2.2639, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 8.989690721649485, | |
| "eval_gen_len": 254.744241, | |
| "eval_loss": 2.3528521060943604, | |
| "eval_rouge1": 0.389675, | |
| "eval_rouge2": 0.083846, | |
| "eval_rougeL": 0.187545, | |
| "eval_rougeLsum": 0.361809, | |
| "eval_runtime": 3507.2655, | |
| "eval_samples_per_second": 0.359, | |
| "eval_steps_per_second": 0.09, | |
| "step": 567 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1260, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.5885960285782016e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |