| { |
| "best_metric": 1.3590947389602661, |
| "best_model_checkpoint": "batch32_1e_5_VED/checkpoint-477", |
| "epoch": 8.91588785046729, |
| "eval_steps": 53, |
| "global_step": 477, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.99, |
| "learning_rate": 9.9e-06, |
| "loss": 5.1065, |
| "step": 53 |
| }, |
| { |
| "epoch": 0.99, |
| "eval_bleu": 0.0, |
| "eval_loss": 3.3200461864471436, |
| "eval_meteor": 0.09540837398436158, |
| "eval_rouge-l": 0.2812177477718752, |
| "eval_runtime": 119.5725, |
| "eval_samples_per_second": 6.331, |
| "eval_steps_per_second": 1.589, |
| "step": 53 |
| }, |
| { |
| "epoch": 1.98, |
| "learning_rate": 9.800000000000001e-06, |
| "loss": 2.8847, |
| "step": 106 |
| }, |
| { |
| "epoch": 1.98, |
| "eval_bleu": 0.012973286002528122, |
| "eval_loss": 2.272148847579956, |
| "eval_meteor": 0.14245785226015273, |
| "eval_rouge-l": 0.2968901288707194, |
| "eval_runtime": 49.2416, |
| "eval_samples_per_second": 15.373, |
| "eval_steps_per_second": 3.859, |
| "step": 106 |
| }, |
| { |
| "epoch": 2.97, |
| "learning_rate": 9.7e-06, |
| "loss": 2.2103, |
| "step": 159 |
| }, |
| { |
| "epoch": 2.97, |
| "eval_bleu": 0.030473187586190942, |
| "eval_loss": 1.9050257205963135, |
| "eval_meteor": 0.1878936938586751, |
| "eval_rouge-l": 0.33559049640535077, |
| "eval_runtime": 49.8376, |
| "eval_samples_per_second": 15.189, |
| "eval_steps_per_second": 3.812, |
| "step": 159 |
| }, |
| { |
| "epoch": 3.96, |
| "learning_rate": 9.600000000000001e-06, |
| "loss": 1.912, |
| "step": 212 |
| }, |
| { |
| "epoch": 3.96, |
| "eval_bleu": 0.04192188158953473, |
| "eval_loss": 1.7162450551986694, |
| "eval_meteor": 0.22255287713312594, |
| "eval_rouge-l": 0.3599682615987815, |
| "eval_runtime": 50.0168, |
| "eval_samples_per_second": 15.135, |
| "eval_steps_per_second": 3.799, |
| "step": 212 |
| }, |
| { |
| "epoch": 4.95, |
| "learning_rate": 9.5e-06, |
| "loss": 1.7459, |
| "step": 265 |
| }, |
| { |
| "epoch": 4.95, |
| "eval_bleu": 0.0490495203639936, |
| "eval_loss": 1.5971415042877197, |
| "eval_meteor": 0.24389844343929945, |
| "eval_rouge-l": 0.37372403459707215, |
| "eval_runtime": 49.4077, |
| "eval_samples_per_second": 15.322, |
| "eval_steps_per_second": 3.846, |
| "step": 265 |
| }, |
| { |
| "epoch": 5.94, |
| "learning_rate": 9.4e-06, |
| "loss": 1.6256, |
| "step": 318 |
| }, |
| { |
| "epoch": 5.94, |
| "eval_bleu": 0.056020605839494524, |
| "eval_loss": 1.5170336961746216, |
| "eval_meteor": 0.25645713891657596, |
| "eval_rouge-l": 0.37995731200625377, |
| "eval_runtime": 50.5532, |
| "eval_samples_per_second": 14.974, |
| "eval_steps_per_second": 3.758, |
| "step": 318 |
| }, |
| { |
| "epoch": 6.93, |
| "learning_rate": 9.3e-06, |
| "loss": 1.533, |
| "step": 371 |
| }, |
| { |
| "epoch": 6.93, |
| "eval_bleu": 0.06304585105562827, |
| "eval_loss": 1.4515591859817505, |
| "eval_meteor": 0.27298702738798486, |
| "eval_rouge-l": 0.38411920753146567, |
| "eval_runtime": 49.065, |
| "eval_samples_per_second": 15.429, |
| "eval_steps_per_second": 3.872, |
| "step": 371 |
| }, |
| { |
| "epoch": 7.93, |
| "learning_rate": 9.200000000000002e-06, |
| "loss": 1.463, |
| "step": 424 |
| }, |
| { |
| "epoch": 7.93, |
| "eval_bleu": 0.06671158142206053, |
| "eval_loss": 1.3997350931167603, |
| "eval_meteor": 0.27980688079271815, |
| "eval_rouge-l": 0.3945394499882324, |
| "eval_runtime": 50.5781, |
| "eval_samples_per_second": 14.967, |
| "eval_steps_per_second": 3.757, |
| "step": 424 |
| }, |
| { |
| "epoch": 8.92, |
| "learning_rate": 9.100000000000001e-06, |
| "loss": 1.4072, |
| "step": 477 |
| }, |
| { |
| "epoch": 8.92, |
| "eval_bleu": 0.07361803483343807, |
| "eval_loss": 1.3590947389602661, |
| "eval_meteor": 0.28844799537676097, |
| "eval_rouge-l": 0.4029055929345029, |
| "eval_runtime": 51.3522, |
| "eval_samples_per_second": 14.741, |
| "eval_steps_per_second": 3.7, |
| "step": 477 |
| } |
| ], |
| "logging_steps": 53, |
| "max_steps": 5300, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 100, |
| "save_steps": 53, |
| "total_flos": 1.0989098432282493e+19, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
