sms1097/utility_model
08
1---2license: mit3datasets:4- sms1097/self_rag_tokens_train_data5---6# Utility Model7 8This generates the `IsUseful` token as descirbed in Self-RAG. 9 10We are testing to see if an answer is useful to the given user question. We output a score from 1-5 based on how useful the answer is. 11 12The expected input to the model is:13 14```15Instruction: {instruction}\nAnswer: {answer}",16```17 18 19### Training Results20```21{'eval_loss': 0.4719298779964447, 22'eval_mse': 0.4719298183917999,23'eval_mae': 0.25655537843704224,24'eval_r2': 0.5200293292355334,25'eval_accuracy': 0.9001516683518705}26```