| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 42500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 6.019460201263428, | |
| "learning_rate": 1.9890916083711463e-05, | |
| "loss": 5.6802, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 6.810026168823242, | |
| "learning_rate": 1.956604419500441e-05, | |
| "loss": 4.6806, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 6.1810994148254395, | |
| "learning_rate": 1.903278917399589e-05, | |
| "loss": 4.2519, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 8.082925796508789, | |
| "learning_rate": 1.8302664407887464e-05, | |
| "loss": 4.0159, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.9830644130706787, | |
| "eval_runtime": 43.0618, | |
| "eval_samples_per_second": 21.945, | |
| "eval_steps_per_second": 10.984, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 8.127954483032227, | |
| "learning_rate": 1.7391582974447918e-05, | |
| "loss": 3.8114, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 7.067851543426514, | |
| "learning_rate": 1.6319401904115078e-05, | |
| "loss": 3.6809, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 5.324134349822998, | |
| "learning_rate": 1.5109489395786475e-05, | |
| "loss": 3.543, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 7.896311283111572, | |
| "learning_rate": 1.3788215506652589e-05, | |
| "loss": 3.4469, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.529372215270996, | |
| "eval_runtime": 42.9645, | |
| "eval_samples_per_second": 21.995, | |
| "eval_steps_per_second": 11.009, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 6.048886299133301, | |
| "learning_rate": 1.2383659531824114e-05, | |
| "loss": 3.3359, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 11.071381568908691, | |
| "learning_rate": 1.0927835785755585e-05, | |
| "loss": 3.2524, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 6.23878288269043, | |
| "learning_rate": 9.452527936105895e-06, | |
| "loss": 3.1945, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 10.52528190612793, | |
| "learning_rate": 7.987692156775495e-06, | |
| "loss": 3.1826, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.3124592304229736, | |
| "eval_runtime": 42.9758, | |
| "eval_samples_per_second": 21.989, | |
| "eval_steps_per_second": 11.006, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 8.30835247039795, | |
| "learning_rate": 6.5674527384518596e-06, | |
| "loss": 3.13, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 9.941985130310059, | |
| "learning_rate": 5.222025727614697e-06, | |
| "loss": 3.0309, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 9.763885498046875, | |
| "learning_rate": 3.980144450189792e-06, | |
| "loss": 2.9941, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 6.858188629150391, | |
| "learning_rate": 2.870115344968626e-06, | |
| "loss": 2.9857, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 6.428621292114258, | |
| "learning_rate": 1.9150471164958117e-06, | |
| "loss": 2.9872, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.2195053100585938, | |
| "eval_runtime": 43.0622, | |
| "eval_samples_per_second": 21.945, | |
| "eval_steps_per_second": 10.984, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 8.133091926574707, | |
| "learning_rate": 1.1370511328570388e-06, | |
| "loss": 2.9005, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 7.232800483703613, | |
| "learning_rate": 5.51546055042601e-07, | |
| "loss": 2.8886, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 7.69180965423584, | |
| "learning_rate": 1.7231245512348337e-07, | |
| "loss": 2.9014, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 11.764738082885742, | |
| "learning_rate": 7.273419542405124e-09, | |
| "loss": 2.9304, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 3.2059476375579834, | |
| "eval_runtime": 42.8633, | |
| "eval_samples_per_second": 22.047, | |
| "eval_steps_per_second": 11.035, | |
| "step": 42500 | |
| } | |
| ], | |
| "logging_steps": 2000, | |
| "max_steps": 42500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 6.876713325662208e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |