Skip to content

Fix per token atrributes bits #7749

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 12 commits into from
Jun 4, 2024
Merged

Fix per token atrributes bits #7749

Changes from all commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
cec6a3b
Add per token attrib enum
Jun 1, 2024
3ead1b9
Using phi-3 for testing 'rstrip'
Jun 1, 2024
33de247
bugfix: assertions, wrong special token list
Jun 1, 2024
ada961c
Implement 'rstrip' properly
Jun 1, 2024
01c9229
Refactor + add 'jina-v2' for testing 'lstrip'
Jun 1, 2024
8564c19
Update phi-3 GGUF file (obsolete since 917dc8c)
Jun 2, 2024
83cac0e
Update brute force test: testing 'lstrip' and 'rstrip'
Jun 2, 2024
54e9f23
Fix previous commit
Jun 2, 2024
ac40ff0
Replace llama_token_type with llama_token_attribs
Jun 2, 2024
18f5fc7
Rename token attributes
Jun 3, 2024
f95c7dc
Fix per token atrributes bits
Jun 4, 2024
6a0f3db
Merge branch 'master' into per-token-attribs
jaime-m-p Jun 4, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 10 additions & 10 deletions llama.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,16 +109,16 @@ extern "C" {

enum llama_token_attr {
LLAMA_TOKEN_ATTR_UNDEFINED = 0,
LLAMA_TOKEN_ATTR_UNKNOWN = 1 << 1,
LLAMA_TOKEN_ATTR_UNUSED = 1 << 2,
LLAMA_TOKEN_ATTR_NORMAL = 1 << 3,
LLAMA_TOKEN_ATTR_CONTROL = 1 << 4, // SPECIAL?
LLAMA_TOKEN_ATTR_USER_DEFINED = 1 << 5,
LLAMA_TOKEN_ATTR_BYTE = 1 << 6,
LLAMA_TOKEN_ATTR_NORMALIZED = 1 << 7,
LLAMA_TOKEN_ATTR_LSTRIP = 1 << 8,
LLAMA_TOKEN_ATTR_RSTRIP = 1 << 9,
LLAMA_TOKEN_ATTR_SINGLE_WORD = 1 << 10,
LLAMA_TOKEN_ATTR_UNKNOWN = 1 << 0,
LLAMA_TOKEN_ATTR_UNUSED = 1 << 1,
LLAMA_TOKEN_ATTR_NORMAL = 1 << 2,
LLAMA_TOKEN_ATTR_CONTROL = 1 << 3, // SPECIAL?
LLAMA_TOKEN_ATTR_USER_DEFINED = 1 << 4,
LLAMA_TOKEN_ATTR_BYTE = 1 << 5,
LLAMA_TOKEN_ATTR_NORMALIZED = 1 << 6,
LLAMA_TOKEN_ATTR_LSTRIP = 1 << 7,
LLAMA_TOKEN_ATTR_RSTRIP = 1 << 8,
LLAMA_TOKEN_ATTR_SINGLE_WORD = 1 << 9,
};

// model file types
Expand Down
Loading