diff options
Diffstat (limited to 'src/llama-vocab.cpp')
| -rw-r--r-- | src/llama-vocab.cpp | 21 |
1 files changed, 21 insertions, 0 deletions
diff --git a/src/llama-vocab.cpp b/src/llama-vocab.cpp index 4bd5aa81..09399417 100644 --- a/src/llama-vocab.cpp +++ b/src/llama-vocab.cpp @@ -439,6 +439,27 @@ struct llm_tokenizer_bpe { "[^\\r\\n\\p{L}\\p{N}]?((?=[\\p{L}])([^a-z]))*((?=[\\p{L}])([^A-Z]))+|[^\\r\\n\\p{L}\\p{N}]?((?=[\\p{L}])([^a-z]))+((?=[\\p{L}])([^A-Z]))*|\\p{N}| ?[^\\s\\p{L}\\p{N}]+[\\r\\n/]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+", }; break; + case LLAMA_VOCAB_PRE_TYPE_GPT4O: + regex_exprs = { + // original regex from tokenizer.json + // "[^\\r\\n\\p{L}\\p{N}]?[\\p{Lu}\\p{Lt}\\p{Lm}\\p{Lo}\\p{M}]*[\\p{Ll}\\p{Lm}\\p{Lo}\\p{M}]+(?i:'s|'t|'re|'ve|'m|'ll|'d)?|[^\\r\\n\\p{L}\\p{N}]?[\\p{Lu}\\p{Lt}\\p{Lm}\\p{Lo}\\p{M}]+[\\p{Ll}\\p{Lm}\\p{Lo}\\p{M}]*(?i:'s|'t|'re|'ve|'m|'ll|'d)?|\\p{N}{1,3}| ?[^\\s\\p{L}\\p{N}]+[\\r\\n/]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+", + "[^\\r\\n\\p{L}\\p{N}]?((?=[\\p{L}])([^a-z]))*((?=[\\p{L}])([^A-Z]))+(?:'[sS]|'[tT]|'[rR][eE]|'[vV][eE]|'[mM]|'[lL][lL]|'[dD])?|[^\\r\\n\\p{L}\\p{N}]?((?=[\\p{L}])([^a-z]))+((?=[\\p{L}])([^A-Z]))*(?:'[sS]|'[tT]|'[rR][eE]|'[vV][eE]|'[mM]|'[lL][lL]|'[dD])?|\\p{N}{1,3}| ?[^\\s\\p{L}\\p{N}]+[\\r\\n/]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+", + }; + break; + case LLAMA_VOCAB_PRE_TYPE_SUPERBPE: + regex_exprs = { + "\\p{N}+", + "(?=(\\d{3})+(?!\\d))", + }; + break; + case LLAMA_VOCAB_PRE_TYPE_BAILINGMOE: + regex_exprs = { + // original regex from tokenizer.json + // "'(?i:[sdmt]|ll|ve|re)|[^\\r\\n\\p{L}\\p{N}]?+\\p{L}+|\\p{N}| ?[^\\s\\p{L}\\p{N}]++[\\r\\n]*|\\s*[\\r\\n]|\\s+(?!\\S)|\\s+" + // FIXME? Changed possessive quantifiers (?+ and ++) to greedy to avoid errors and imatrix hanging (tried atomic grouping but it's not supported?) + "'(?:[sSdDmMtT]|[lL][lL]|[vV][eE]|[rR][eE])|[^\\r\\n\\p{L}\\p{N}]?\\p{L}+|\\p{N}| ?[^\\s\\p{L}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]|\\s+(?!\\S)|\\s+", + }; + break; default: // default regex for BPE tokenization pre-processing regex_exprs = { |