| { | |
| "best_metric": 0.5598750842730023, | |
| "best_model_checkpoint": "results/checkpoint-3500", | |
| "epoch": 0.7, | |
| "global_step": 3500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 6.6e-05, | |
| "loss": 2.0145, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_bleu": 0.1150567678154011, | |
| "eval_exact_match": 0.132, | |
| "eval_loss": 0.9738793969154358, | |
| "eval_rouge1": 0.5382577445488361, | |
| "eval_rouge2": 0.3545728830144713, | |
| "eval_rougeL": 0.5187026156468759, | |
| "eval_rougeLsum": 0.5259109026044895, | |
| "eval_runtime": 59.0107, | |
| "eval_samples_per_second": 33.892, | |
| "eval_steps_per_second": 1.695, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.828070175438598e-05, | |
| "loss": 0.8761, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_bleu": 0.11272954562081616, | |
| "eval_exact_match": 0.1475, | |
| "eval_loss": 0.901577353477478, | |
| "eval_rouge1": 0.5359795989196405, | |
| "eval_rouge2": 0.3649742450836393, | |
| "eval_rougeL": 0.5187273882783475, | |
| "eval_rougeLsum": 0.5253745777261007, | |
| "eval_runtime": 55.4587, | |
| "eval_samples_per_second": 36.063, | |
| "eval_steps_per_second": 1.803, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.477192982456141e-05, | |
| "loss": 0.8139, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_bleu": 0.11562105292517781, | |
| "eval_exact_match": 0.1425, | |
| "eval_loss": 0.8843896389007568, | |
| "eval_rouge1": 0.5385440798738839, | |
| "eval_rouge2": 0.3630877932160531, | |
| "eval_rougeL": 0.520238931080043, | |
| "eval_rougeLsum": 0.5274836357515698, | |
| "eval_runtime": 55.2097, | |
| "eval_samples_per_second": 36.226, | |
| "eval_steps_per_second": 1.811, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.126315789473685e-05, | |
| "loss": 0.7745, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_bleu": 0.11999069408174769, | |
| "eval_exact_match": 0.152, | |
| "eval_loss": 0.8501775860786438, | |
| "eval_rouge1": 0.5495869647995661, | |
| "eval_rouge2": 0.37385624817390106, | |
| "eval_rougeL": 0.5307934771920211, | |
| "eval_rougeLsum": 0.5391651788581161, | |
| "eval_runtime": 56.8684, | |
| "eval_samples_per_second": 35.169, | |
| "eval_steps_per_second": 1.758, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 8.77543859649123e-05, | |
| "loss": 0.7416, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_bleu": 0.12262269562147479, | |
| "eval_exact_match": 0.148, | |
| "eval_loss": 0.8417201042175293, | |
| "eval_rouge1": 0.5524556612739725, | |
| "eval_rouge2": 0.37743352139147385, | |
| "eval_rougeL": 0.5333960691072436, | |
| "eval_rougeLsum": 0.5408174364513855, | |
| "eval_runtime": 55.5707, | |
| "eval_samples_per_second": 35.99, | |
| "eval_steps_per_second": 1.8, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.424561403508773e-05, | |
| "loss": 0.7097, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_bleu": 0.12307160579590622, | |
| "eval_exact_match": 0.155, | |
| "eval_loss": 0.8275034427642822, | |
| "eval_rouge1": 0.5571591325212851, | |
| "eval_rouge2": 0.3846841785043018, | |
| "eval_rougeL": 0.5390722174679519, | |
| "eval_rougeLsum": 0.5456202385734745, | |
| "eval_runtime": 56.9486, | |
| "eval_samples_per_second": 35.119, | |
| "eval_steps_per_second": 1.756, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 8.073684210526316e-05, | |
| "loss": 0.6923, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_bleu": 0.1269094917224047, | |
| "eval_exact_match": 0.156, | |
| "eval_loss": 0.8161399960517883, | |
| "eval_rouge1": 0.5598750842730023, | |
| "eval_rouge2": 0.38392351740583797, | |
| "eval_rougeL": 0.5409288136858783, | |
| "eval_rougeLsum": 0.5484962105303921, | |
| "eval_runtime": 55.0573, | |
| "eval_samples_per_second": 36.326, | |
| "eval_steps_per_second": 1.816, | |
| "step": 3500 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 3, | |
| "total_flos": 8985722848542720.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |