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

Rwkv chat template fix #10001

Merged
merged 3 commits into from
Oct 22, 2024
Merged

Rwkv chat template fix #10001

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
543b102
llama: remove useless template matching for rwkv-world
MollySophia Oct 22, 2024
facec46
converter: Add comment about the hack for rwkv models
MollySophia Oct 22, 2024
1135b0f
Update src/llama.cpp
MollySophia Oct 22, 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
1 change: 1 addition & 0 deletions convert_hf_to_gguf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2865,6 +2865,7 @@ def set_vocab(self):
self.gguf_writer.add_token_types(toktypes)
special_vocab = gguf.SpecialVocab(self.dir_model, load_merges=False)
special_vocab.chat_template = "rwkv-world"
# hack: Add '\n\n' as the EOT token to make it chat normally
special_vocab._set_special_token("eot", 261)
special_vocab.add_to_gguf(self.gguf_writer)

Expand Down
3 changes: 2 additions & 1 deletion src/llama.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21697,7 +21697,8 @@ static int32_t llama_chat_apply_template_internal(
if (add_ass) {
ss << "[|assistant|]";
}
} else if (tmpl == "rwkv-world" || tmpl_contains("rwkv-world") || tmpl_contains("'User: ' + message['content'] + '\n\nAssistant:'")) {
} else if (tmpl == "rwkv-world" || tmpl_contains("rwkv-world")) {
MollySophia marked this conversation as resolved.
Show resolved Hide resolved
// this template requires the model to have "\n\n" as EOT token
for (auto message : chat) {
std::string role(message->role);
if (role == "user") {
Expand Down
4 changes: 0 additions & 4 deletions tests/test-chat-template.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,8 +65,6 @@ int main(void) {
u8"{% for message in messages %}{% if message['role'] == 'user' %}{{'<用户>' + message['content'].strip() + '<AI>'}}{% else %}{{message['content'].strip()}}{% endif %}{% endfor %}",
// DeepSeek-V2
"{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{{ bos_token }}{% for message in messages %}{% if message['role'] == 'user' %}{{ 'User: ' + message['content'] + '\n\n' }}{% elif message['role'] == 'assistant' %}{{ 'Assistant: ' + message['content'] + eos_token }}{% elif message['role'] == 'system' %}{{ message['content'] + '\n\n' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'Assistant:' }}{% endif %}",
// RWKV-World
"{% for message in messages %}{% if message['role'] == 'user' %}{{'User: ' + message['content'] + '\n\nAssistant:'}}{% else %}{{message['content'] + '\n\n'}}{% endif %}{% endfor %}",
};
std::vector<std::string> expected_output = {
// teknium/OpenHermes-2.5-Mistral-7B
Expand Down Expand Up @@ -111,8 +109,6 @@ int main(void) {
u8"You are a helpful assistant<用户>Hello<AI>Hi there<用户>Who are you<AI>I am an assistant<用户>Another question<AI>",
// DeepSeek-V2
u8"You are a helpful assistant\n\nUser: Hello\n\nAssistant: Hi there<｜end▁of▁sentence｜>User: Who are you\n\nAssistant: I am an assistant <｜end▁of▁sentence｜>User: Another question\n\nAssistant:",
// RWKV-World
"You are a helpful assistant\n\nUser: Hello\n\nAssistant:Hi there\n\nUser: Who are you\n\nAssistant: I am an assistant \n\nUser: Another question\n\nAssistant:",
};
std::vector<char> formatted_chat(1024);
int32_t res;
Expand Down
Loading