stanford-crfm
/

marin-tokenizer

Model card Files Files and versions

dlwh commited on Jun 13

Commit

bed5584

·

verified ·

1 Parent(s): d53740f

Upload tokenizer

Files changed (2) hide show

tokenizer.json +2 -2
tokenizer_config.json +2 -2

tokenizer.json CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
-size 17209920

 version https://git-lfs.github.com/spec/v1
+oid sha256:881c9c36c359e1617afef6f7583403567931b7b4f43f6552d2b2155a131650a2
+size 17209892

tokenizer_config.json CHANGED Viewed

@@ -17,7 +17,7 @@
       "special": true
     },
     "128002": {
-      "content": "<|reserved_special_token_0|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -25,7 +25,7 @@
       "special": true
     },
     "128003": {
-      "content": "<|reserved_special_token_1|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,

       "special": true
     },
     "128002": {
+      "content": "<|start_think|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "special": true
     },
     "128003": {
+      "content": "<|end_think|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,