| { | |
| "best_metric": 0.07522870600223541, | |
| "best_model_checkpoint": "./teapotllm/checkpoint-1640", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 8200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.528669357299805, | |
| "learning_rate": 4.75e-05, | |
| "loss": 6.5278, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.08746226876974106, | |
| "eval_runtime": 4.8581, | |
| "eval_samples_per_second": 45.285, | |
| "eval_steps_per_second": 5.764, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.16966140270233154, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0895, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.0768546611070633, | |
| "eval_runtime": 4.8636, | |
| "eval_samples_per_second": 45.234, | |
| "eval_steps_per_second": 5.757, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.40588390827178955, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.1199, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.07700426131486893, | |
| "eval_runtime": 4.8598, | |
| "eval_samples_per_second": 45.269, | |
| "eval_steps_per_second": 5.762, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.0932304635643959, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0639, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.07522870600223541, | |
| "eval_runtime": 4.8695, | |
| "eval_samples_per_second": 45.179, | |
| "eval_steps_per_second": 5.75, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.08160164952278137, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0692, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.07598432898521423, | |
| "eval_runtime": 4.8511, | |
| "eval_samples_per_second": 45.351, | |
| "eval_steps_per_second": 5.772, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.24587534368038177, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.061, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.0796995759010315, | |
| "eval_runtime": 4.8601, | |
| "eval_samples_per_second": 45.266, | |
| "eval_steps_per_second": 5.761, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.5349581837654114, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.0395, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.07927663624286652, | |
| "eval_runtime": 4.8456, | |
| "eval_samples_per_second": 45.402, | |
| "eval_steps_per_second": 5.778, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.19880425930023193, | |
| "learning_rate": 3e-05, | |
| "loss": 0.036, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.0807080939412117, | |
| "eval_runtime": 4.8744, | |
| "eval_samples_per_second": 45.134, | |
| "eval_steps_per_second": 5.744, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.10939527302980423, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.0345, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.08234285563230515, | |
| "eval_runtime": 4.863, | |
| "eval_samples_per_second": 45.24, | |
| "eval_steps_per_second": 5.758, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.45931077003479004, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0303, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.08353324979543686, | |
| "eval_runtime": 4.8563, | |
| "eval_samples_per_second": 45.302, | |
| "eval_steps_per_second": 5.766, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 43.76845932006836, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.0267, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.08770459145307541, | |
| "eval_runtime": 4.8587, | |
| "eval_samples_per_second": 45.28, | |
| "eval_steps_per_second": 5.763, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.8835374116897583, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0243, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.08723071962594986, | |
| "eval_runtime": 4.867, | |
| "eval_samples_per_second": 45.203, | |
| "eval_steps_per_second": 5.753, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.07083321362733841, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.0215, | |
| "step": 5330 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.09219814836978912, | |
| "eval_runtime": 4.8773, | |
| "eval_samples_per_second": 45.107, | |
| "eval_steps_per_second": 5.741, | |
| "step": 5330 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.1330474615097046, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0206, | |
| "step": 5740 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.09269961714744568, | |
| "eval_runtime": 4.8707, | |
| "eval_samples_per_second": 45.168, | |
| "eval_steps_per_second": 5.749, | |
| "step": 5740 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.20611341297626495, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0193, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.09489532560110092, | |
| "eval_runtime": 4.8553, | |
| "eval_samples_per_second": 45.311, | |
| "eval_steps_per_second": 5.767, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 24.089698791503906, | |
| "learning_rate": 1e-05, | |
| "loss": 0.019, | |
| "step": 6560 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.09438970685005188, | |
| "eval_runtime": 4.8722, | |
| "eval_samples_per_second": 45.154, | |
| "eval_steps_per_second": 5.747, | |
| "step": 6560 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.29387062788009644, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.018, | |
| "step": 6970 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.097468800842762, | |
| "eval_runtime": 4.8614, | |
| "eval_samples_per_second": 45.254, | |
| "eval_steps_per_second": 5.76, | |
| "step": 6970 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.06152157112956047, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0174, | |
| "step": 7380 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.09819656610488892, | |
| "eval_runtime": 4.8553, | |
| "eval_samples_per_second": 45.312, | |
| "eval_steps_per_second": 5.767, | |
| "step": 7380 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.20427298545837402, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0674, | |
| "step": 7790 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.09810814261436462, | |
| "eval_runtime": 4.8519, | |
| "eval_samples_per_second": 45.343, | |
| "eval_steps_per_second": 5.771, | |
| "step": 7790 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 12.638993263244629, | |
| "learning_rate": 0.0, | |
| "loss": 0.0171, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.0987108051776886, | |
| "eval_runtime": 4.8753, | |
| "eval_samples_per_second": 45.126, | |
| "eval_steps_per_second": 5.743, | |
| "step": 8200 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.12300205801472e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |