| { | |
| "best_global_step": 100, | |
| "best_metric": 0.6652931571006775, | |
| "best_model_checkpoint": "./luau-model/checkpoint-100", | |
| "epoch": 3.0, | |
| "eval_steps": 100, | |
| "global_step": 117, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1282051282051282, | |
| "grad_norm": 0.3991023898124695, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.1619, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2564102564102564, | |
| "grad_norm": 0.5307971239089966, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.1662, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 0.6414325833320618, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.1415, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5128205128205128, | |
| "grad_norm": 4.2611846923828125, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.9051, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6410256410256411, | |
| "grad_norm": 0.38137564063072205, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8326, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.32487988471984863, | |
| "learning_rate": 0.000116, | |
| "loss": 0.7638, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8974358974358975, | |
| "grad_norm": 0.39361336827278137, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 0.7069, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.0256410256410255, | |
| "grad_norm": 0.3810354471206665, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 0.652, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 0.33373090624809265, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 0.651, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.282051282051282, | |
| "grad_norm": 0.46736040711402893, | |
| "learning_rate": 0.000196, | |
| "loss": 0.6554, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4102564102564101, | |
| "grad_norm": 0.34294310212135315, | |
| "learning_rate": 0.00018805970149253734, | |
| "loss": 0.557, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.36410510540008545, | |
| "learning_rate": 0.00017313432835820895, | |
| "loss": 0.5899, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.8159669041633606, | |
| "learning_rate": 0.00015820895522388059, | |
| "loss": 0.5576, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.7948717948717947, | |
| "grad_norm": 0.41050034761428833, | |
| "learning_rate": 0.00014328358208955225, | |
| "loss": 0.5889, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 0.3205547034740448, | |
| "learning_rate": 0.00012835820895522389, | |
| "loss": 0.5561, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "grad_norm": 0.29004010558128357, | |
| "learning_rate": 0.00011343283582089552, | |
| "loss": 0.5052, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.1794871794871793, | |
| "grad_norm": 0.40997835993766785, | |
| "learning_rate": 9.850746268656717e-05, | |
| "loss": 0.4612, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 0.3654220700263977, | |
| "learning_rate": 8.358208955223881e-05, | |
| "loss": 0.4086, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.435897435897436, | |
| "grad_norm": 0.34712859988212585, | |
| "learning_rate": 6.865671641791044e-05, | |
| "loss": 0.4658, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "grad_norm": 0.4480563700199127, | |
| "learning_rate": 5.373134328358209e-05, | |
| "loss": 0.4577, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "eval_loss": 0.6652931571006775, | |
| "eval_runtime": 5.1656, | |
| "eval_samples_per_second": 10.067, | |
| "eval_steps_per_second": 1.742, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 0.47177863121032715, | |
| "learning_rate": 3.8805970149253736e-05, | |
| "loss": 0.4419, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.8205128205128203, | |
| "grad_norm": 0.4423898160457611, | |
| "learning_rate": 2.3880597014925373e-05, | |
| "loss": 0.3984, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.948717948717949, | |
| "grad_norm": 0.4147056043148041, | |
| "learning_rate": 8.955223880597016e-06, | |
| "loss": 0.3729, | |
| "step": 115 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 117, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.859666246520013e+16, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |