datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
bielik-q2-sharp
Bielik Q2# Research: Evaluation Results & Documentation
Evaluation results, scripts, logs, and reports from 2-bit quantization
research on speakleash/Bielik-11B-v2.3-Instruct.
Variants
Variant A: QuIP# E8P12 (successful)
Method: QuIP# with E8P12 lattice codebook, 2-bit
Model: Jakubrd4/Bielik-11B-v2.3-Instruct-QuIP-2bit
Size: 3.26 GB (vs ~22 GB FP16, ~6.7x compression)
Normalized avg (22 tasks): 61.10 (vs 65.71 FP16, ~93% retention)
Evaluation: Full Polish LLM… See the full description on the dataset page: https://huggingface.co/datasets/Jakubrd4/bielik-q2-sharp.details_speakleash__Bielik-11B-v2.3-Instruct
Dataset Card for Evaluation run of speakleash/Bielik-11B-v2.3-Instruct
Dataset automatically created during the evaluation run of model speakleash/Bielik-11B-v2.3-Instruct.
The dataset is composed of 136 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 latest… See the full description on the dataset page: https://huggingface.co/datasets/OALL/details_speakleash__Bielik-11B-v2.3-Instruct.details_speakleash__Bielik-7B-v0.1
Dataset Card for Evaluation run of speakleash/Bielik-7B-v0.1
Dataset automatically created during the evaluation run of model speakleash/Bielik-7B-v0.1 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 2 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_speakleash__Bielik-7B-v0.1.details_TeeZee__Bielik-SOLAR-LIKE-10.7B-Instruct-v0.1
Dataset Card for Evaluation run of TeeZee/Bielik-SOLAR-LIKE-10.7B-Instruct-v0.1
Dataset automatically created during the evaluation run of model TeeZee/Bielik-SOLAR-LIKE-10.7B-Instruct-v0.1 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_TeeZee__Bielik-SOLAR-LIKE-10.7B-Instruct-v0.1.bielik-distill-polish-10k
bielik-distill-polish-10k
Polish instruction-tuning dataset with 10,304 samples generated via response-level knowledge distillation from Bielik-11B-v3.0-Instruct (SpeakLeash, Apache 2.0).
Covers Polish history, culture, politics, science, geography, idioms, and general reasoning. Multi-pass quality control: factual corrections, topic filtering (Poland/Europe focus), truncation removal (~9% of raw data removed).
Format
{
"messages": [
{"role": "user"… See the full description on the dataset page: https://huggingface.co/datasets/JohnTdi/bielik-distill-polish-10k.speakleash__Bielik-11B-v2.3-Instruct-details
Dataset Card for Evaluation run of speakleash/Bielik-11B-v2.3-Instruct
Dataset automatically created during the evaluation run of model speakleash/Bielik-11B-v2.3-Instruct
The dataset is composed of 38 configuration(s), each one corresponding 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 latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/speakleash__Bielik-11B-v2.3-Instruct-details.details_speakleash__Bielik-7B-Instruct-v0.1
Dataset Card for Evaluation run of speakleash/Bielik-7B-Instruct-v0.1
Dataset automatically created during the evaluation run of model speakleash/Bielik-7B-Instruct-v0.1 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 2 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_speakleash__Bielik-7B-Instruct-v0.1.bielik-q2-sharp-docsspeakleash__Bielik-11B-v2-details
Dataset Card for Evaluation run of speakleash/Bielik-11B-v2
Dataset automatically created during the evaluation run of model speakleash/Bielik-11B-v2
The dataset is composed of 38 configuration(s), each one corresponding 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 latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/speakleash__Bielik-11B-v2-details.speakleash__Bielik-11B-v2.1-Instruct-detailsspeakleash__Bielik-11B-v2.0-Instruct-details
Dataset Card for Evaluation run of speakleash/Bielik-11B-v2.0-Instruct
Dataset automatically created during the evaluation run of model speakleash/Bielik-11B-v2.0-Instruct
The dataset is composed of 38 configuration(s), each one corresponding 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 latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/speakleash__Bielik-11B-v2.0-Instruct-details.speakleash__Bielik-11B-v2.3-InstructBielikowo
Datasets for Bielikowo tutorials
Bielikowo Github
speakleash__Bielik-11B-v2.2-Instruct-detailsspeakleash__Bielik-11B-v2.0-Instructmeditsolutions__MSH-v1-Bielik-v2.3-Instruct-MedIT-mergemeditsolutions__MSH-v1-Bielik-v2.3-Instruct-MedIT-merge-details
Dataset Card for Evaluation run of meditsolutions/MSH-v1-Bielik-v2.3-Instruct-MedIT-merge
Dataset automatically created during the evaluation run of model meditsolutions/MSH-v1-Bielik-v2.3-Instruct-MedIT-merge
The dataset is composed of 38 configuration(s), each one corresponding 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/meditsolutions__MSH-v1-Bielik-v2.3-Instruct-MedIT-merge-details.details_speakleash__Bielik-7B-v0.1
Dataset Card for Evaluation run of speakleash/Bielik-7B-v0.1
Dataset automatically created during the evaluation run of model speakleash/Bielik-7B-v0.1 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 2 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/Studio2kPolska/details_speakleash__Bielik-7B-v0.1.details_speakleash__Bielik-11B-v2
Dataset Card for Evaluation run of speakleash/Bielik-11B-v2
Dataset automatically created during the evaluation run of model speakleash/Bielik-11B-v2.
The dataset is composed of 136 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 latest results.
An additional… See the full description on the dataset page: https://huggingface.co/datasets/OALL/details_speakleash__Bielik-11B-v2.meditsolutions__MSH-Lite-7B-v1-Bielik-v2.3-Instruct-Llama-Prune-details
Dataset Card for Evaluation run of meditsolutions/MSH-Lite-7B-v1-Bielik-v2.3-Instruct-Llama-Prune
Dataset automatically created during the evaluation run of model meditsolutions/MSH-Lite-7B-v1-Bielik-v2.3-Instruct-Llama-Prune
The dataset is composed of 38 configuration(s), each one corresponding 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/meditsolutions__MSH-Lite-7B-v1-Bielik-v2.3-Instruct-Llama-Prune-details.slayer-v49-qwen3.5-27b-human-pref-v49-vs-bielik
v49 vs Bielik v3 11B Blind Human Preference Set
This dataset contains 100 diverse prompts with two anonymized model answers
per prompt. Human judges should compare answer_a and answer_b without knowing
which model produced each answer.
Suggested judging labels
A: answer A is better
B: answer B is better
tie: both are about equally good
bad_both: both answers are unacceptable
Judge on helpfulness, correctness, completeness, instruction following, and
clarity. Do… See the full description on the dataset page: https://huggingface.co/datasets/kacperwikiel/slayer-v49-qwen3.5-27b-human-pref-v49-vs-bielik.meditsolutions__MSH-Lite-7B-v1-Bielik-v2.3-Instruct-Llama-Prunebielik-q2-hessians