Update README.md

README.md

@@ -1,22 +1,12 @@
 ---
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-dataset_info:
-  features:
-  - name: text
-    dtype: string
-  - name: embeddings
-    sequence: float32
-  splits:
-  - name: train
-    num_bytes: 46275561696
-    num_examples: 5489000
-  download_size: 20572356672
-  dataset_size: 46275561696
+# For reference on model card metadata, see the spec: https://github.com/huggingface/hub-docs/blob/main/datasetcard.md?plain=1
+# Doc / guide: https://huggingface.co/docs/hub/datasets-cards
+{}
 ---
-# Dataset Card for "SlimPajama-6B-embedded"
 
-[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
+# Dataset Card for SlimPajama-6B-embedded
+
+This is a copy of [DKYoon/SlimPajama-6B](https://huggingface.co/datasets/DKYoon/SlimPajama-6B), together with embeddings generated by [thenlper/gte-large](https://huggingface.co/thenlper/gte-large). 
+There are 5.49 million examples of text, a representative random sample of [SlimPajama-627B](https://huggingface.co/datasets/cerebras/SlimPajama-627B). Each text is associated with a 1024-dimensional embedding vector that is meant to represent the semantic content. The vectors were generated by average-pooling (max-pooling dataset to come in the future).
+
+This dataset is intended to help with downstream tasks such as reverse-embeddings, interpreting embedding spaces and creating adapters between embeddings models.