5a-community/vdr_jina_v4_layernorm_data

vdr_jina_v4_layernorm_data This dataset contains precomputed internal-layer text and image embeddings with jina-embedding-v4 Instead of storing raw text/image as primary training inputs, each row stores layer-wise embedding vectors: text_layer_00, text_layer_01, ..., text_layer_NN image_layer_00, image_layer_01, ..., image_layer_NN query_is_empty (boolean indicator used during filtering/evaluation) These artifacts are intended for probing, head training, and retrieval… See the full description on the dataset page: https://huggingface.co/datasets/5a-community/vdr_jina_v4_layernorm_data.

<iframe src="https://silytics.si/embed/item/925eecd9-e2fc-4485-8fe2-891464eee76a" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.