Upload triples.train.hard_negatives.tsv

Ranked the top 200 passages using cosine similarity between query and passage embeddings in msmarco's qrels.train.tsv generated by sentence-transformers/all-MiniLM-L6-v2. Then randomly sampled a single passage whose rank 'r' is 50<=r<=200 to get the hard negative for each query.

.gitattributes

@@ -57,3 +57,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+triples.train.hard_negatives.tsv filter=lfs diff=lfs merge=lfs -text

triples.train.hard_negatives.tsv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:284806235f4c79a977ab48aa605d1a92871af330589ed169e8ba4de4a0075414
+size 367457127