| { | |
| "best_metric": 1.531933307647705, | |
| "best_model_checkpoint": "./tst-translation-output/checkpoint-1500", | |
| "epoch": 5.609573672400898, | |
| "eval_steps": 1500, | |
| "global_step": 7500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7074, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.952812382030955e-05, | |
| "loss": 1.6178, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.9056247640619104e-05, | |
| "loss": 1.4354, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_bleu": 11.2991, | |
| "eval_gen_len": 16.268, | |
| "eval_loss": 1.531933307647705, | |
| "eval_runtime": 168.1897, | |
| "eval_samples_per_second": 15.887, | |
| "eval_steps_per_second": 0.993, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.8584371460928654e-05, | |
| "loss": 1.1627, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 4.8112495281238205e-05, | |
| "loss": 1.1969, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 4.7640619101547756e-05, | |
| "loss": 0.9158, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_bleu": 11.9085, | |
| "eval_gen_len": 16.3331, | |
| "eval_loss": 1.5989255905151367, | |
| "eval_runtime": 174.1685, | |
| "eval_samples_per_second": 15.341, | |
| "eval_steps_per_second": 0.959, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.7168742921857306e-05, | |
| "loss": 0.7923, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.669686674216686e-05, | |
| "loss": 0.831, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 4.622499056247641e-05, | |
| "loss": 0.4823, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "eval_bleu": 11.2452, | |
| "eval_gen_len": 15.771, | |
| "eval_loss": 1.7406657934188843, | |
| "eval_runtime": 159.7198, | |
| "eval_samples_per_second": 16.729, | |
| "eval_steps_per_second": 1.046, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 4.575311438278596e-05, | |
| "loss": 0.5185, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 4.528123820309551e-05, | |
| "loss": 0.4567, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 4.480936202340506e-05, | |
| "loss": 0.3005, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "eval_bleu": 11.0872, | |
| "eval_gen_len": 16.3668, | |
| "eval_loss": 1.8923007249832153, | |
| "eval_runtime": 162.6697, | |
| "eval_samples_per_second": 16.426, | |
| "eval_steps_per_second": 1.027, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 4.433748584371461e-05, | |
| "loss": 0.3253, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 4.386560966402416e-05, | |
| "loss": 0.2367, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 4.339373348433372e-05, | |
| "loss": 0.1969, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "eval_bleu": 11.449, | |
| "eval_gen_len": 15.8694, | |
| "eval_loss": 2.0163686275482178, | |
| "eval_runtime": 160.1148, | |
| "eval_samples_per_second": 16.688, | |
| "eval_steps_per_second": 1.043, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "step": 7500, | |
| "total_flos": 2.6005557215232e+17, | |
| "train_loss": 0.8117534647623698, | |
| "train_runtime": 8858.9284, | |
| "train_samples_per_second": 96.54, | |
| "train_steps_per_second": 6.037 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 53480, | |
| "num_train_epochs": 40, | |
| "save_steps": 1500, | |
| "total_flos": 2.6005557215232e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |