Skip to content
This repository has been archived by the owner on Aug 30, 2024. It is now read-only.

[Model Validation] Support Mistral-base-v0.2 #192

Merged
merged 3 commits into from
Mar 27, 2024
Merged

[Model Validation] Support Mistral-base-v0.2 #192

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
fdb9211
update doc
Zhenzhong1 Mar 25, 2024
47cd81f
fixed the script issue
Zhenzhong1 Mar 25, 2024
0ec8399
add the total number of paraemters
Zhenzhong1 Mar 25, 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
13 changes: 6 additions & 7 deletions docs/supported_models.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -94,6 +94,7 @@ Neural Speed supports the following models:
</tr>
<tr>
<td><a href="https://huggingface.co/mistralai/Mistral-7B-v0.1" target="_blank" rel="noopener noreferrer">Mistral-7B</a>,
<a href="https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2" target="_blank" rel="noopener noreferrer">Mistral-7B-Instruct-v0.2</a>,
<a href="https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1" target="_blank" rel="noopener noreferrer">Mixtral-8x7B</a></td>
<td>✅</td>
<td>✅</td>
Expand Down Expand Up @@ -402,15 +403,15 @@ Neural Speed supports the following models:
<td></td>
</tr>
<tr>
<td><a href="https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF" target="_blank" rel="noopener noreferrer">TheBloke/Mistral-7B-v0.1-GGUF</a>,
<td><a href="https://huggingface.co/TheBloke/Mistral-7B-v0.1-GGUF" target="_blank" rel="noopener noreferrer">TheBloke/Mistral-7B-v0.1-GGUF</a>, <a href="https://huggingface.co/TheBloke/Mistral-7B-v0.2-GGUF" target="_blank" rel="noopener noreferrer">TheBloke/Mistral-7B-v0.2-GGUF</a>,
<td>✅</td>
<td>✅</td>
<td>✅</td>
<td>✅</td>
<td></td>
</tr>
<tr>
<td><a href="https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF" target="_blank" rel="noopener noreferrer">TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUFF</a>,
<td><a href="https://huggingface.co/TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF" target="_blank" rel="noopener noreferrer">TheBloke/Mixtral-8x7B-Instruct-v0.1-GGUF</a>
<td>✅</td>
<td>✅</td>
<td>✅</td>
Expand All @@ -425,18 +426,16 @@ Neural Speed supports the following models:
<td>✅</td>
<td></td>
</tr>
</tr>
<tr>
<td><a href="https://huggingface.co/codellama/CodeLlama-7b-hf" target="_blank" rel="noopener noreferrer">TheBloke/CodeLlama-7B-GGUF</a></td>
<td><a href="https://huggingface.co/codellama/CodeLlama-7b-hf" target="_blank" rel="noopener noreferrer">TheBloke/CodeLlama-7B-GGUF</a>,<a href="https://huggingface.co/codellama/CodeLlama-13b-hf" target="_blank" rel="noopener noreferrer">TheBloke/CodeLlama-13B-GGUF</a></td>
<td>✅</td>
<td>✅</td>
<td>✅</td>
<td>✅</td>
<td></td>
</tr>
</tr>
<tr>
<td><a href="https://huggingface.co/codellama/CodeLlama-13b-hf" target="_blank" rel="noopener noreferrer">TheBloke/CodeLlama-13B-GGUF</a></td>
<td><a href="https://huggingface.co/Qwen/Qwen1.5-7B-Chat-GGUF" target="_blank" rel="noopener noreferrer">Qwen1.5-7B-Chat-GGUF</a></td>
<td>✅</td>
<td>✅</td>
<td>✅</td>
Expand Down Expand Up @@ -470,7 +469,7 @@ Neural Speed supports the following models:
</tr>
<tr>
<td><a href="https://huggingface.co/Qwen/Qwen-7B-Chat" target="_blank" rel="noopener noreferrer">Qwen-7B-Chat</a>,
<a href="https://huggingface.co/Qwen/Qwen1.5-7B-Chat-GGUF" target="_blank" rel="noopener noreferrer">Qwen1.5-7B-Chat-GGUF</a></td>
<a href="https://huggingface.co/Qwen/Qwen1.5-7B-Chat" target="_blank" rel="noopener noreferrer">Qwen1.5-7B-Chat</a></td>
<td>✅</td>
<td>✅</td>
<td>✅</td>
Expand Down
6 changes: 3 additions & 3 deletions neural_speed/models/model_utils/model_files.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1138,15 +1138,15 @@ struct model_file_loader {
printf("%-16s %d.hparams.original_max_position_embeddings = %-30d\n", __func__, count++,
hparams.original_max_position_embeddings);
printf("%-16s %d.hparams.use_yarn = %-30d\n", __func__, count++, hparams.use_yarn);
unsigned int total = 25;
unsigned int total = 26;
if (count != total) {
fprintf(stderr, "The number of ne_parameters is wrong.\n");
fprintf(stderr, "The number of ne_parameters is wrong, total = %d, count = %d.\n", total, count);
}
}

void load_ne_vocab() {
unsigned int count = 0;
unsigned int ne_hparams_total = 25;
unsigned int ne_hparams_total = 26;
file.read_raw(&vocab.bos_token_id, sizeof(model_vocab::id));
file.read_raw(&vocab.eos_token_id, sizeof(model_vocab::id));
file.read_raw(&vocab.pad_token_id, sizeof(model_vocab::id));
Expand Down
2 changes: 1 addition & 1 deletion scripts/python_api_example_for_gguf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,7 +50,7 @@ def main(args_in: Optional[List[str]] = None) -> None:

gguf_path = args.model.as_posix()

prompt = "Once upon a time"
prompt = args.prompt
tokenizer = AutoTokenizer.from_pretrained(args.model_path, trust_remote_code=True)
inputs = tokenizer(prompt, return_tensors="pt").input_ids
streamer = TextStreamer(tokenizer)
Expand Down
Loading