model: handle multiple eos tokens (#10577)

* get eos_token_id from generation_config.json

* refactor

* include both ids and strings in trace

* comments

* remove special case for gemma3 special vocab (#10743)
This commit is contained in:
Michael Yang
2025-05-16 13:40:23 -07:00
committed by GitHub
parent 27da2cddc5
commit 333e360422
18 changed files with 282 additions and 182 deletions

View File

@@ -110,6 +110,7 @@ func parseTokenizer(fsys fs.FS, specialTokenTypes []string) (*Tokenizer, error)
}
if f, err := fsys.Open("tokenizer_config.json"); errors.Is(err, os.ErrNotExist) {
// noop
} else if err != nil {
return nil, err
} else {
@@ -171,6 +172,34 @@ func parseTokenizer(fsys fs.FS, specialTokenTypes []string) (*Tokenizer, error)
}
}
if f, err := fsys.Open("generation_config.json"); errors.Is(err, os.ErrNotExist) {
} else if err != nil {
return nil, err
} else {
defer f.Close()
var p map[string]json.RawMessage
if err := json.NewDecoder(f).Decode(&p); err != nil {
return nil, err
}
for _, st := range specialTokenTypes {
if bts, ok := p[fmt.Sprintf("%s_token_id", st)]; ok {
var ids []int32
if err := json.Unmarshal(bts, &ids); err != nil {
// value is not a list so the existing ID is used
continue
}
if i := slices.IndexFunc(t.SpecialVocabulary, func(sv *SpecialVocabulary) bool {
return sv.Type == st
}); i >= 0 {
t.SpecialVocabulary[i].IDs = ids
}
}
}
}
return t, nil
}
@@ -280,6 +309,9 @@ type SpecialVocabulary struct {
ID int
Content string
AddToken bool
// IDs is populated by generation_config.json
IDs []int32
}
func (sv SpecialVocabulary) Key() string {