6 references to EndOfText
Microsoft.ML.Tokenizers (6)
Model\TiktokenTokenizer.cs (6)
1124{ { EndOfText, 100257}, { FimPrefix, 100258}, { FimMiddle, 100259}, { FimSuffix, 100260}, { EndOfPrompt, 100276} }, Cl100kBaseRegex(), Cl100kBaseVocabFile, Type.GetType(Cl100kBaseTypeName), Cl100kBasePackageName);
1127return (new Dictionary<string, int> { { EndOfText, 50256 }, }, P50kBaseRegex(), GPT2File, Type.GetType(Gpt2TypeName), Gpt2PackageName);
1130return (new Dictionary<string, int> { { EndOfText, 199999 }, { EndOfPrompt, 200018 } }, O200kBaseRegex(), O200kBaseFile, Type.GetType(O200kBaseTypeName), O200kBasePackageName);
1133return (new Dictionary<string, int> { { EndOfText, 50256 } }, P50kBaseRegex(), P50RanksFile, Type.GetType(P50kBaseTypeName), P50kBasePackageName);
1137{ { EndOfText, 50256 }, { FimPrefix, 50281 }, { FimMiddle, 50282 }, { FimSuffix, 50283 } }, P50kBaseRegex(), P50RanksFile, Type.GetType(P50kBaseTypeName), P50kBasePackageName);
1140return (new Dictionary<string, int> { { EndOfText, 50256 } }, P50kBaseRegex(), R50RanksFile, Type.GetType(R50kBaseTypeName), R50kBasePackageName);