| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.028169014084507, | |
| "eval_steps": 54, | |
| "global_step": 432, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 2.4982516765594482, | |
| "learning_rate": 1.267605633802817e-05, | |
| "loss": 1.266, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.2101283073425293, | |
| "eval_runtime": 55.9111, | |
| "eval_samples_per_second": 4.024, | |
| "eval_steps_per_second": 1.019, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 2.3082025051116943, | |
| "learning_rate": 2.535211267605634e-05, | |
| "loss": 1.0833, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 0.9703609347343445, | |
| "eval_runtime": 55.946, | |
| "eval_samples_per_second": 4.022, | |
| "eval_steps_per_second": 1.019, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 2.2696995735168457, | |
| "learning_rate": 3.802816901408451e-05, | |
| "loss": 0.9416, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.9111042022705078, | |
| "eval_runtime": 55.9589, | |
| "eval_samples_per_second": 4.021, | |
| "eval_steps_per_second": 1.019, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 2.649714469909668, | |
| "learning_rate": 4.992175273865415e-05, | |
| "loss": 0.9319, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.8902803659439087, | |
| "eval_runtime": 55.9001, | |
| "eval_samples_per_second": 4.025, | |
| "eval_steps_per_second": 1.02, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 2.1477463245391846, | |
| "learning_rate": 4.85133020344288e-05, | |
| "loss": 0.8269, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 0.8803441524505615, | |
| "eval_runtime": 55.901, | |
| "eval_samples_per_second": 4.025, | |
| "eval_steps_per_second": 1.02, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 2.4527807235717773, | |
| "learning_rate": 4.710485133020345e-05, | |
| "loss": 0.8594, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 0.8677402138710022, | |
| "eval_runtime": 55.8955, | |
| "eval_samples_per_second": 4.025, | |
| "eval_steps_per_second": 1.02, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 2.6781976222991943, | |
| "learning_rate": 4.569640062597809e-05, | |
| "loss": 0.881, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_loss": 0.8588340878486633, | |
| "eval_runtime": 55.8965, | |
| "eval_samples_per_second": 4.025, | |
| "eval_steps_per_second": 1.02, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 3.1084585189819336, | |
| "learning_rate": 4.428794992175274e-05, | |
| "loss": 0.8246, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_loss": 0.8576346635818481, | |
| "eval_runtime": 55.889, | |
| "eval_samples_per_second": 4.026, | |
| "eval_steps_per_second": 1.02, | |
| "step": 432 | |
| } | |
| ], | |
| "logging_steps": 54, | |
| "max_steps": 2130, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 54, | |
| "total_flos": 3.273170558356685e+16, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |