| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6298815822625347, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.5e-05, | |
| "loss": 37.0618, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 27.398916244506836, | |
| "eval_runtime": 3.1349, | |
| "eval_samples_per_second": 31.899, | |
| "eval_steps_per_second": 4.147, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9e-05, | |
| "loss": 34.5311, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 36.926673889160156, | |
| "eval_runtime": 3.1612, | |
| "eval_samples_per_second": 31.633, | |
| "eval_steps_per_second": 4.112, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.5e-05, | |
| "loss": 33.5869, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 31.76376724243164, | |
| "eval_runtime": 3.1795, | |
| "eval_samples_per_second": 31.452, | |
| "eval_steps_per_second": 4.089, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 8e-05, | |
| "loss": 34.2002, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 33.623844146728516, | |
| "eval_runtime": 3.1768, | |
| "eval_samples_per_second": 31.479, | |
| "eval_steps_per_second": 4.092, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 34.0832, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 27.767799377441406, | |
| "eval_runtime": 3.166, | |
| "eval_samples_per_second": 31.586, | |
| "eval_steps_per_second": 4.106, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 7e-05, | |
| "loss": 32.413, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 34.710330963134766, | |
| "eval_runtime": 3.1812, | |
| "eval_samples_per_second": 31.434, | |
| "eval_steps_per_second": 4.086, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 33.3394, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 29.091354370117188, | |
| "eval_runtime": 3.1601, | |
| "eval_samples_per_second": 31.645, | |
| "eval_steps_per_second": 4.114, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 6e-05, | |
| "loss": 32.6119, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 30.3073673248291, | |
| "eval_runtime": 3.1555, | |
| "eval_samples_per_second": 31.691, | |
| "eval_steps_per_second": 4.12, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 33.4702, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 30.2010555267334, | |
| "eval_runtime": 3.1576, | |
| "eval_samples_per_second": 31.67, | |
| "eval_steps_per_second": 4.117, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 5e-05, | |
| "loss": 32.9516, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 28.628616333007812, | |
| "eval_runtime": 3.1678, | |
| "eval_samples_per_second": 31.568, | |
| "eval_steps_per_second": 4.104, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10000, | |
| "num_train_epochs": 2, | |
| "save_steps": 5000, | |
| "total_flos": 6.84757352448e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |