| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5119672340970178, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0004957330602491893, | |
| "loss": 2.0441, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0004914661204983786, | |
| "loss": 1.713, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00048719918074756785, | |
| "loss": 1.6007, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00048293224099675715, | |
| "loss": 1.5181, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00047866530124594645, | |
| "loss": 1.4493, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_bleu": 3.9997520454971136, | |
| "eval_loss": 1.3540712594985962, | |
| "eval_runtime": 10.3322, | |
| "eval_samples_per_second": 193.569, | |
| "eval_steps_per_second": 1.549, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0004743983614951357, | |
| "loss": 1.4096, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.000470131421744325, | |
| "loss": 1.3729, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0004658644819935143, | |
| "loss": 1.3389, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00046159754224270354, | |
| "loss": 1.3166, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00045733060249189284, | |
| "loss": 1.2915, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_bleu": 5.393565315342523, | |
| "eval_loss": 1.2113043069839478, | |
| "eval_runtime": 8.5631, | |
| "eval_samples_per_second": 233.562, | |
| "eval_steps_per_second": 1.868, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00045306366274108214, | |
| "loss": 1.2669, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0004487967229902714, | |
| "loss": 1.2492, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0004445297832394607, | |
| "loss": 1.2371, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00044026284348864993, | |
| "loss": 1.2231, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00043599590373783923, | |
| "loss": 1.2059, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_bleu": 5.838067918496751, | |
| "eval_loss": 1.1367273330688477, | |
| "eval_runtime": 9.2404, | |
| "eval_samples_per_second": 216.442, | |
| "eval_steps_per_second": 1.732, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0004317289639870285, | |
| "loss": 1.1931, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0004274620242362178, | |
| "loss": 1.1761, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0004231950844854071, | |
| "loss": 1.1792, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0004189281447345963, | |
| "loss": 1.1614, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0004146612049837856, | |
| "loss": 1.1573, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_bleu": 6.242227124133382, | |
| "eval_loss": 1.0900604724884033, | |
| "eval_runtime": 8.3953, | |
| "eval_samples_per_second": 238.228, | |
| "eval_steps_per_second": 1.906, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 11718, | |
| "num_train_epochs": 3, | |
| "save_steps": 2000, | |
| "total_flos": 5883781806292992.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |