|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 10.0, |
|
"eval_steps": 500, |
|
"global_step": 5290, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 0.0005, |
|
"loss": 1.3631, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 0.0005, |
|
"loss": 1.294, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 0.0005, |
|
"loss": 1.302, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 0.0005, |
|
"loss": 1.2626, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 0.0005, |
|
"loss": 1.2156, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.6911294117647059, |
|
"eval_loss": 1.012805700302124, |
|
"eval_runtime": 7.0918, |
|
"eval_samples_per_second": 70.504, |
|
"eval_steps_per_second": 8.883, |
|
"step": 529 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_exact_match": 27.6, |
|
"eval_f1": 35.68145743145743, |
|
"eval_qa_bleu": 19.064111689202868, |
|
"eval_qa_exact_match": 0.216, |
|
"eval_recite_bleu": 17.898525975872754, |
|
"eval_recite_exact_match": 0.006, |
|
"step": 529 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"learning_rate": 0.0005, |
|
"loss": 1.0305, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.32, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9199, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 0.0005, |
|
"loss": 0.9086, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"learning_rate": 0.0005, |
|
"loss": 2.9245, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 0.0005, |
|
"loss": 4.9199, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.22922352941176471, |
|
"eval_loss": 7.1224870681762695, |
|
"eval_runtime": 7.7159, |
|
"eval_samples_per_second": 64.801, |
|
"eval_steps_per_second": 8.165, |
|
"step": 1058 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 1058 |
|
}, |
|
{ |
|
"epoch": 2.08, |
|
"learning_rate": 0.0005, |
|
"loss": 7.4349, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"learning_rate": 0.0005, |
|
"loss": 8.0921, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 2.46, |
|
"learning_rate": 0.0005, |
|
"loss": 7.889, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 2.65, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2559, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3563, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.24235294117647058, |
|
"eval_loss": 7.308440685272217, |
|
"eval_runtime": 7.7087, |
|
"eval_samples_per_second": 64.862, |
|
"eval_steps_per_second": 8.173, |
|
"step": 1587 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 1587 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2721, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 3.21, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2743, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 3.4, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3151, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 3.59, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3366, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 3.78, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3294, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 3.97, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3158, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.23965490196078432, |
|
"eval_loss": 7.242833614349365, |
|
"eval_runtime": 6.6551, |
|
"eval_samples_per_second": 75.13, |
|
"eval_steps_per_second": 9.466, |
|
"step": 2116 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 2116 |
|
}, |
|
{ |
|
"epoch": 4.16, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3307, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 4.35, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3307, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 4.54, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3197, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 4.73, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3238, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 4.91, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3017, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.24474509803921568, |
|
"eval_loss": 7.229503631591797, |
|
"eval_runtime": 7.0755, |
|
"eval_samples_per_second": 70.667, |
|
"eval_steps_per_second": 8.904, |
|
"step": 2645 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 2645 |
|
}, |
|
{ |
|
"epoch": 5.1, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2926, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 5.29, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3006, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 5.48, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3051, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 5.67, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2973, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 5.86, |
|
"learning_rate": 0.0005, |
|
"loss": 7.281, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.24474509803921568, |
|
"eval_loss": 7.229503631591797, |
|
"eval_runtime": 7.1043, |
|
"eval_samples_per_second": 70.38, |
|
"eval_steps_per_second": 8.868, |
|
"step": 3174 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 3174 |
|
}, |
|
{ |
|
"epoch": 6.05, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3017, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 6.24, |
|
"learning_rate": 0.0005, |
|
"loss": 7.297, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 6.43, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2877, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 6.62, |
|
"learning_rate": 0.0005, |
|
"loss": 7.296, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 6.81, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2993, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3056, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.24474509803921568, |
|
"eval_loss": 7.229503631591797, |
|
"eval_runtime": 7.6931, |
|
"eval_samples_per_second": 64.993, |
|
"eval_steps_per_second": 8.189, |
|
"step": 3703 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 3703 |
|
}, |
|
{ |
|
"epoch": 7.18, |
|
"learning_rate": 0.0005, |
|
"loss": 7.299, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 7.37, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2954, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 7.56, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3002, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 7.75, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3011, |
|
"step": 4100 |
|
}, |
|
{ |
|
"epoch": 7.94, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2943, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.24474509803921568, |
|
"eval_loss": 7.229503631591797, |
|
"eval_runtime": 6.6465, |
|
"eval_samples_per_second": 75.228, |
|
"eval_steps_per_second": 9.479, |
|
"step": 4232 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 4232 |
|
}, |
|
{ |
|
"epoch": 8.13, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3001, |
|
"step": 4300 |
|
}, |
|
{ |
|
"epoch": 8.32, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2922, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 8.51, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2969, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 8.7, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2879, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 8.88, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3011, |
|
"step": 4700 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.24474509803921568, |
|
"eval_loss": 7.229503631591797, |
|
"eval_runtime": 7.77, |
|
"eval_samples_per_second": 64.35, |
|
"eval_steps_per_second": 8.108, |
|
"step": 4761 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 4761 |
|
}, |
|
{ |
|
"epoch": 9.07, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3066, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 9.26, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2949, |
|
"step": 4900 |
|
}, |
|
{ |
|
"epoch": 9.45, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2955, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 9.64, |
|
"learning_rate": 0.0005, |
|
"loss": 7.3061, |
|
"step": 5100 |
|
}, |
|
{ |
|
"epoch": 9.83, |
|
"learning_rate": 0.0005, |
|
"loss": 7.2948, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.24474509803921568, |
|
"eval_loss": 7.229503631591797, |
|
"eval_runtime": 7.0663, |
|
"eval_samples_per_second": 70.758, |
|
"eval_steps_per_second": 8.916, |
|
"step": 5290 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_exact_match": 0.0, |
|
"eval_f1": 0.0, |
|
"eval_qa_bleu": 0.0, |
|
"eval_qa_exact_match": 0.0, |
|
"eval_recite_bleu": 0.0, |
|
"eval_recite_exact_match": 0.0, |
|
"step": 5290 |
|
} |
|
], |
|
"logging_steps": 100, |
|
"max_steps": 26450, |
|
"num_train_epochs": 50, |
|
"save_steps": 500, |
|
"total_flos": 1.1216445557880914e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|