| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0003, | |
| "loss": 1.9296, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0003, | |
| "loss": 1.7554, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6093164556962025, | |
| "eval_loss": 1.7939746379852295, | |
| "eval_runtime": 5.0177, | |
| "eval_samples_per_second": 99.648, | |
| "eval_steps_per_second": 12.556, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 26.0, | |
| "eval_f1": 36.71136363636365, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0003, | |
| "loss": 1.6394, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0003, | |
| "loss": 1.5315, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0003, | |
| "loss": 1.5248, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6085063291139241, | |
| "eval_loss": 1.8273799419403076, | |
| "eval_runtime": 4.4023, | |
| "eval_samples_per_second": 113.576, | |
| "eval_steps_per_second": 14.311, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 24.4, | |
| "eval_f1": 33.57723554223556, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0003, | |
| "loss": 1.1692, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0003, | |
| "loss": 1.2054, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6027088607594937, | |
| "eval_loss": 1.9717934131622314, | |
| "eval_runtime": 5.1273, | |
| "eval_samples_per_second": 97.518, | |
| "eval_steps_per_second": 12.287, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 20.8, | |
| "eval_f1": 30.584243697478996, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0003, | |
| "loss": 1.0252, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8608, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0003, | |
| "loss": 0.8989, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5986835443037974, | |
| "eval_loss": 2.151914596557617, | |
| "eval_runtime": 5.3325, | |
| "eval_samples_per_second": 93.765, | |
| "eval_steps_per_second": 11.814, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 21.0, | |
| "eval_f1": 30.363229548229558, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5842, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0003, | |
| "loss": 0.6306, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5960506329113924, | |
| "eval_loss": 2.329284191131592, | |
| "eval_runtime": 4.5585, | |
| "eval_samples_per_second": 109.686, | |
| "eval_steps_per_second": 13.82, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 20.4, | |
| "eval_f1": 30.02542624042625, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 0.0003, | |
| "loss": 0.5368, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.0003, | |
| "loss": 0.4495, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.0003, | |
| "loss": 0.4712, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5935696202531645, | |
| "eval_loss": 2.5598793029785156, | |
| "eval_runtime": 4.4094, | |
| "eval_samples_per_second": 113.395, | |
| "eval_steps_per_second": 14.288, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 20.6, | |
| "eval_f1": 30.108149073149082, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.0003, | |
| "loss": 0.3625, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 0.0003, | |
| "loss": 0.3797, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5935696202531645, | |
| "eval_loss": 2.732860803604126, | |
| "eval_runtime": 5.2701, | |
| "eval_samples_per_second": 94.876, | |
| "eval_steps_per_second": 11.954, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_exact_match": 19.6, | |
| "eval_f1": 29.002709512709526, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.0003, | |
| "loss": 0.3582, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 0.0003, | |
| "loss": 0.3399, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.0003, | |
| "loss": 0.3527, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5912911392405064, | |
| "eval_loss": 2.8185083866119385, | |
| "eval_runtime": 4.4105, | |
| "eval_samples_per_second": 113.366, | |
| "eval_steps_per_second": 14.284, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_exact_match": 20.8, | |
| "eval_f1": 29.191525696525694, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12500, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.2781970097648435e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |