mistral7b
Mistral-7B-v0.1-base-tokenized-dolma-v1_7-50Bmistral-7b-hidden-states-tpu-verification
Mistral-7B TPU hidden-state extractor verification
Verification-only artifact — not a reproduction of the paper's AUROC.
This dataset contains last-token embeddings and hidden states extracted from
1,000 flattened CoQA validation question/reference-answer pairs with
mistralai/Mistral-7B-Instruct-v0.3. It verifies that the memory-bounded TPU
v5e-1 extraction path runs successfully. It does not contain the Mistral
best_answer generations or hallucination labels required to… See the full description on the dataset page: https://huggingface.co/datasets/GwendalTsang/mistral-7b-hidden-states-tpu-verification.Mistral-7B-LongPO-256K-tokenizedMistral-7B-LongPO-128K-tokenizeddetails_xxyyy123__Mistral7B_adaptor_v1
Dataset Card for Evaluation run of xxyyy123/Mistral7B_adaptor_v1
Dataset Summary
Dataset automatically created during the evaluation run of model xxyyy123/Mistral7B_adaptor_v1 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"… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_xxyyy123__Mistral7B_adaptor_v1.details_RefalMachine__ruadapt_mistral7b_full_vo_1e4