self-rag
SciPhi-Self-RAG-Mistral-7B-32k-GGUFselfrag_llama2_7bSciPhi-Self-RAG-Mistral-7B-32k-Mistral-7B-Instruct-v0.2-slerp-GGUFSciPhi-Self-RAG-Mistral-7B-32k-i1-GGUFSciPhi-Self-RAG-Mistral-7B-32k-Mistral-7B-Instruct-v0.1-GGUFselfrag_llama2_7b-GGUFSciPhi-Self-RAG-Mistral-7B-32k-GGUFSciPhi-Self-RAG-Mistral-7B-32k
Datasets
All datasets matching “self-rag”crag_self_rag_datasetsselfrag_train_dataThis is a training data file for Self-RAG that generates outputs to diverse user queries as well as reflection tokens to call the retrieval system adaptively and criticize its own output and retrieved passages.
Self-RAG is trained on our 150k diverse instruction-output pairs with interleaving passages and reflection tokens using the standard next-token prediction objective, enabling efficient and stable learning with fine-grained feedback.
At inference, we leverage reflection tokens covering… See the full description on the dataset page: https://huggingface.co/datasets/selfrag/selfrag_train_data.details_selfrag__selfrag_llama2_7b
Dataset Card for Evaluation run of selfrag/selfrag_llama2_7b
Dataset automatically created during the evaluation run of model selfrag/selfrag_llama2_7b on the Open LLM Leaderboard.
The dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_selfrag__selfrag_llama2_7b.details_SciPhi__SciPhi-Self-RAG-Mistral-7B-32k
Dataset Card for Evaluation run of SciPhi/SciPhi-Self-RAG-Mistral-7B-32k
Dataset Summary
Dataset automatically created during the evaluation run of model SciPhi/SciPhi-Self-RAG-Mistral-7B-32k on the Open LLM Leaderboard.
The dataset is composed of 64 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_SciPhi__SciPhi-Self-RAG-Mistral-7B-32k.selfrag-eval-dataself_rag_tokens_train_data
Self-Rag Tokens Dataset
This dataset is a spin off of the work from Self-Rag training data.
In Self-RAG, the authors show how a LLM can be trained to predict tokens for retrieval, if the context is relevant/irrelevant, if the answer is supported, and how useful the response is.
The limitation of Self-RAG is that you must train the LLM on this task, which can be tricky or cost prohibitive. With rapid developments in LLM performance, investing in training one LLM with Self-RAG may… See the full description on the dataset page: https://huggingface.co/datasets/sms1097/self_rag_tokens_train_data.