datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
audio_interference_mmlummlu_speech
MMLU Speech
Speech version of MMLU eval, where the speech is synthesized using XTTS-v2. Note that there might not be a 1:1 mapping with the original text eval due to TTS failures.
MMLU_FULL-emilia_5s_dsmmlu_speech
MMLU Speech
Speech version of MMLU eval, where the speech is synthesized using XTTS-v2. Note that there might not be a 1:1 mapping with the original text eval due to TTS failures.
MMLU_FULL-full_train_hf_format.merged.v1speech_mmlu_demointerleaving_mmlu_demointerleaving_mmlummlummlu_eval_audioaudio_mmlu_biologyaudio_mmlu_high_school_biologymmlu_eval_resultinterleaving_mmlu_deprecatedinterleaving_mmlu_nofilterspeech_mmlu_deprecatedMMLU_FULL-full_train_dpo_hf_format.catm2ram2.v1MMLU_FULL-full_5k_hf_format.v1