diff_llama
diffllama_patch_tokenizeddetails_undi95__llama2-to-mistral-diff
Dataset Card for Evaluation run of undi95/llama2-to-mistral-diff
Dataset Summary
Dataset automatically created during the evaluation run of model undi95/llama2-to-mistral-diff 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 3 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_undi95__llama2-to-mistral-diff.qwen3_llama3_rank_diff_k6
qwen3_llama3_rank_diff_k6
This repository contains k-means cluster subsets for training. Each subset is a config (name=cluster_X), backed by Parquet.
Load examples
from datasets import load_dataset
# Single cluster
ds0 = load_dataset("{username}/qwen3_llama3_rank_diff_k6", name="cluster_0")
# Specific cluster
ds3 = load_dataset("{username}/qwen3_llama3_rank_diff_k6", name="cluster_3")
# Or with data_files directly (optional alternative):
ds =… See the full description on the dataset page: https://huggingface.co/datasets/Adanato/qwen3_llama3_rank_diff_k6.crosscoder-llama-3.2-1b-diffcrosscoder-llama-3.2-1b-diffllama31_test_chat_format_20k_only_firstwrong_and_regular_first_corr10k_diff_prompt_ep3tmp0