Skip to content
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

Jump to bottom

llama : Added support for SmolLm pre-tokenizer (#8608) #8609

Merged
merged 5 commits into from
Jul 22, 2024
Merged

llama : Added support for SmolLm pre-tokenizer (#8608) #8609

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
28bd56f
Adding SmolLM Pre Tokenizer
Stillerman Jul 20, 2024
7647916
Update convert_hf_to_gguf_update.py
Stillerman Jul 21, 2024
689e38c
Update src/llama.cpp
Stillerman Jul 21, 2024
f4600e6
handle regex
Stillerman Jul 21, 2024
525e789
removed .inp and out .out ggufs
Stillerman Jul 21, 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
3 changes: 3 additions & 0 deletions convert_hf_to_gguf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -596,6 +596,9 @@ def get_vocab_base_pre(self, tokenizer) -> str:
if chkhsh == "63b97e4253352e6f357cc59ea5b583e3a680eaeaf2632188c2b952de2588485e":
# ref: https://huggingface.co/mistralai/Mistral-Nemo-Base-2407
res = "tekken"
if chkhsh == "855059429035d75a914d1eda9f10a876752e281a054a7a3d421ef0533e5b6249":
# ref: https://huggingface.co/HuggingFaceTB/SmolLM-135M
res = "smollm"

if res is None:
logger.warning("\n")
Expand Down
1 change: 1 addition & 0 deletions convert_hf_to_gguf_update.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,6 +92,7 @@ class TOKENIZER_TYPE(IntEnum):
{"name": "jais", "tokt": TOKENIZER_TYPE.BPE, "repo": "https://huggingface.co/core42/jais-13b", },
{"name": "t5", "tokt": TOKENIZER_TYPE.UGM, "repo": "https://huggingface.co/google-t5/t5-small", },
{"name": "tekken", "tokt": TOKENIZER_TYPE.BPE, "repo": "https://huggingface.co/mistralai/Mistral-Nemo-Base-2407", },
{"name": "smollm", "tokt": TOKENIZER_TYPE.BPE, "repo": "https://huggingface.co/HuggingFaceTB/SmolLM-135M", },
]


Expand Down
1 change: 1 addition & 0 deletions include/llama.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -93,6 +93,7 @@ extern "C" {
LLAMA_VOCAB_PRE_TYPE_VIKING = 18,
LLAMA_VOCAB_PRE_TYPE_JAIS = 19,
LLAMA_VOCAB_PRE_TYPE_TEKKEN = 20,
LLAMA_VOCAB_PRE_TYPE_SMOLLM = 21,
};

// note: these values should be synchronized with ggml_rope
Expand Down
5 changes: 5 additions & 0 deletions src/llama.cpp
View file
Open in desktop
Stillerman marked this conversation as resolved.
Show resolved Hide resolved
Original file line number Diff line number Diff line change
Expand Up @@ -5530,6 +5530,10 @@ static void llm_load_vocab(
vocab.tokenizer_clean_spaces = false;
vocab.tokenizer_ignore_merges = true;
vocab.tokenizer_add_bos = true;
} else if (
tokenizer_pre == "smollm") {
vocab.type_pre = LLAMA_VOCAB_PRE_TYPE_SMOLLM;
Stillerman marked this conversation as resolved.
Show resolved Hide resolved
vocab.tokenizer_clean_spaces = false;
} else {
throw std::runtime_error(format("unknown pre-tokenizer type: '%s'", tokenizer_pre.c_str()));
}
Expand Down Expand Up @@ -15554,6 +15558,7 @@ struct llm_tokenizer_bpe {
case LLAMA_VOCAB_PRE_TYPE_STARCODER:
case LLAMA_VOCAB_PRE_TYPE_REFACT:
case LLAMA_VOCAB_PRE_TYPE_COMMAND_R:
case LLAMA_VOCAB_PRE_TYPE_SMOLLM:
regex_exprs = {
"\\p{N}",
"'s|'t|'re|'ve|'m|'ll|'d| ?\\p{L}+| ?\\p{N}+| ?[^\\s\\p{L}\\p{N}]+|\\s+(?!\\S)",
Expand Down
Loading