| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.3929173693086, | |
| "global_step": 11500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 9.578414839797639e-06, | |
| "loss": 0.3136, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 17.1673, | |
| "eval_gen_len": 43.5832, | |
| "eval_loss": 2.7004430294036865, | |
| "eval_runtime": 263.1603, | |
| "eval_samples_per_second": 3.792, | |
| "eval_steps_per_second": 0.475, | |
| "step": 593 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 9.156829679595279e-06, | |
| "loss": 0.2962, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 17.2012, | |
| "eval_gen_len": 43.6693, | |
| "eval_loss": 2.719139814376831, | |
| "eval_runtime": 263.247, | |
| "eval_samples_per_second": 3.791, | |
| "eval_steps_per_second": 0.475, | |
| "step": 1186 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 8.735244519392918e-06, | |
| "loss": 0.2927, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 17.2291, | |
| "eval_gen_len": 43.482, | |
| "eval_loss": 2.7412936687469482, | |
| "eval_runtime": 260.7955, | |
| "eval_samples_per_second": 3.827, | |
| "eval_steps_per_second": 0.479, | |
| "step": 1779 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 8.313659359190556e-06, | |
| "loss": 0.2677, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 17.135, | |
| "eval_gen_len": 43.5862, | |
| "eval_loss": 2.7617862224578857, | |
| "eval_runtime": 261.1749, | |
| "eval_samples_per_second": 3.821, | |
| "eval_steps_per_second": 0.479, | |
| "step": 2372 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 7.892074198988196e-06, | |
| "loss": 0.2591, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 17.5543, | |
| "eval_gen_len": 43.5922, | |
| "eval_loss": 2.7780115604400635, | |
| "eval_runtime": 262.647, | |
| "eval_samples_per_second": 3.8, | |
| "eval_steps_per_second": 0.476, | |
| "step": 2965 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 7.470489038785835e-06, | |
| "loss": 0.2473, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.9, | |
| "learning_rate": 7.048903878583474e-06, | |
| "loss": 0.2282, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 17.226, | |
| "eval_gen_len": 43.6703, | |
| "eval_loss": 2.794311761856079, | |
| "eval_runtime": 263.8826, | |
| "eval_samples_per_second": 3.782, | |
| "eval_steps_per_second": 0.474, | |
| "step": 3558 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 6.6273187183811136e-06, | |
| "loss": 0.2244, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 17.615, | |
| "eval_gen_len": 43.6934, | |
| "eval_loss": 2.808680295944214, | |
| "eval_runtime": 264.0527, | |
| "eval_samples_per_second": 3.78, | |
| "eval_steps_per_second": 0.473, | |
| "step": 4151 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 6.2057335581787524e-06, | |
| "loss": 0.2196, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 17.3227, | |
| "eval_gen_len": 43.7715, | |
| "eval_loss": 2.825133800506592, | |
| "eval_runtime": 281.974, | |
| "eval_samples_per_second": 3.539, | |
| "eval_steps_per_second": 0.443, | |
| "step": 4744 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "learning_rate": 5.784148397976391e-06, | |
| "loss": 0.2101, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 17.5072, | |
| "eval_gen_len": 43.7084, | |
| "eval_loss": 2.834676742553711, | |
| "eval_runtime": 263.2231, | |
| "eval_samples_per_second": 3.791, | |
| "eval_steps_per_second": 0.475, | |
| "step": 5337 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 5.362563237774031e-06, | |
| "loss": 0.2077, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 17.5712, | |
| "eval_gen_len": 43.8597, | |
| "eval_loss": 2.842376708984375, | |
| "eval_runtime": 270.4686, | |
| "eval_samples_per_second": 3.69, | |
| "eval_steps_per_second": 0.462, | |
| "step": 5930 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 4.94097807757167e-06, | |
| "loss": 0.2034, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "learning_rate": 4.519392917369309e-06, | |
| "loss": 0.1968, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bleu": 17.6007, | |
| "eval_gen_len": 43.6994, | |
| "eval_loss": 2.851884365081787, | |
| "eval_runtime": 261.9767, | |
| "eval_samples_per_second": 3.809, | |
| "eval_steps_per_second": 0.477, | |
| "step": 6523 | |
| }, | |
| { | |
| "epoch": 11.8, | |
| "learning_rate": 4.097807757166948e-06, | |
| "loss": 0.1902, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_bleu": 17.6333, | |
| "eval_gen_len": 43.6924, | |
| "eval_loss": 2.8614132404327393, | |
| "eval_runtime": 263.2972, | |
| "eval_samples_per_second": 3.79, | |
| "eval_steps_per_second": 0.475, | |
| "step": 7116 | |
| }, | |
| { | |
| "epoch": 12.65, | |
| "learning_rate": 3.676222596964587e-06, | |
| "loss": 0.198, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_bleu": 17.6153, | |
| "eval_gen_len": 43.7034, | |
| "eval_loss": 2.865877866744995, | |
| "eval_runtime": 261.219, | |
| "eval_samples_per_second": 3.821, | |
| "eval_steps_per_second": 0.479, | |
| "step": 7709 | |
| }, | |
| { | |
| "epoch": 13.49, | |
| "learning_rate": 3.2546374367622263e-06, | |
| "loss": 0.1861, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_bleu": 17.5959, | |
| "eval_gen_len": 43.7154, | |
| "eval_loss": 2.873347043991089, | |
| "eval_runtime": 260.1505, | |
| "eval_samples_per_second": 3.836, | |
| "eval_steps_per_second": 0.48, | |
| "step": 8302 | |
| }, | |
| { | |
| "epoch": 14.33, | |
| "learning_rate": 2.8330522765598656e-06, | |
| "loss": 0.1956, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_bleu": 17.6169, | |
| "eval_gen_len": 43.7164, | |
| "eval_loss": 2.876323938369751, | |
| "eval_runtime": 261.1714, | |
| "eval_samples_per_second": 3.821, | |
| "eval_steps_per_second": 0.479, | |
| "step": 8895 | |
| }, | |
| { | |
| "epoch": 15.18, | |
| "learning_rate": 2.4114671163575045e-06, | |
| "loss": 0.1924, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_bleu": 17.5443, | |
| "eval_gen_len": 43.7194, | |
| "eval_loss": 2.880269765853882, | |
| "eval_runtime": 261.8101, | |
| "eval_samples_per_second": 3.812, | |
| "eval_steps_per_second": 0.477, | |
| "step": 9488 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "learning_rate": 1.9898819561551434e-06, | |
| "loss": 0.1946, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 16.86, | |
| "learning_rate": 1.5682967959527825e-06, | |
| "loss": 0.1946, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_bleu": 17.577, | |
| "eval_gen_len": 43.6142, | |
| "eval_loss": 2.8834807872772217, | |
| "eval_runtime": 259.2401, | |
| "eval_samples_per_second": 3.85, | |
| "eval_steps_per_second": 0.482, | |
| "step": 10081 | |
| }, | |
| { | |
| "epoch": 17.71, | |
| "learning_rate": 1.1467116357504218e-06, | |
| "loss": 0.1987, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_bleu": 17.5677, | |
| "eval_gen_len": 43.6623, | |
| "eval_loss": 2.8818464279174805, | |
| "eval_runtime": 269.8962, | |
| "eval_samples_per_second": 3.698, | |
| "eval_steps_per_second": 0.463, | |
| "step": 10674 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 7.251264755480608e-07, | |
| "loss": 0.2011, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_bleu": 17.6118, | |
| "eval_gen_len": 43.7395, | |
| "eval_loss": 2.882765531539917, | |
| "eval_runtime": 265.3175, | |
| "eval_samples_per_second": 3.762, | |
| "eval_steps_per_second": 0.471, | |
| "step": 11267 | |
| }, | |
| { | |
| "epoch": 19.39, | |
| "learning_rate": 3.0354131534569986e-07, | |
| "loss": 0.2049, | |
| "step": 11500 | |
| } | |
| ], | |
| "max_steps": 11860, | |
| "num_train_epochs": 20, | |
| "total_flos": 1623592378957824.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |