[LLM] Small fix to MPT Example (#8513)
This commit is contained in:
parent
ba0da17b40
commit
bcde8ec83e
2 changed files with 4 additions and 4 deletions
|
|
@ -16,7 +16,7 @@ conda create -n llm python=3.9
|
||||||
conda activate llm
|
conda activate llm
|
||||||
|
|
||||||
pip install bigdl-llm[all] # install bigdl-llm with 'all' option
|
pip install bigdl-llm[all] # install bigdl-llm with 'all' option
|
||||||
pip instll einops # additional package required for mpt-7b-chat to conduct generation
|
pip install einops # additional package required for mpt-7b-chat to conduct generation
|
||||||
```
|
```
|
||||||
|
|
||||||
### 2. Config
|
### 2. Config
|
||||||
|
|
|
||||||
|
|
@ -22,14 +22,14 @@ from bigdl.llm.transformers import AutoModelForCausalLM
|
||||||
from transformers import AutoTokenizer
|
from transformers import AutoTokenizer
|
||||||
|
|
||||||
# you could revise it based on the MPT model you choose to use
|
# you could revise it based on the MPT model you choose to use
|
||||||
MPT_PROMPT_FORMAT="<human>{prompt} <bot>"
|
MPT_PROMPT_FORMAT = "<human>{prompt} <bot>"
|
||||||
|
|
||||||
if __name__ == '__main__':
|
if __name__ == '__main__':
|
||||||
parser = argparse.ArgumentParser(description='Transformer INT4 example for MPT model')
|
parser = argparse.ArgumentParser(description='Transformer INT4 example for MPT model')
|
||||||
parser.add_argument('--repo-id-or-model-path', type=str, default="mosaicml/mpt-7b-chat",
|
parser.add_argument('--repo-id-or-model-path', type=str, default="mosaicml/mpt-7b-chat",
|
||||||
help='The huggingface repo id for the large language model to be downloaded'
|
help='The huggingface repo id for the MPT to be downloaded'
|
||||||
', or the path to the huggingface checkpoint folder')
|
', or the path to the huggingface checkpoint folder')
|
||||||
parser.add_argument('--prompt', type=str, default="<human>What is AI?<bot>",
|
parser.add_argument('--prompt', type=str, default="What is AI?",
|
||||||
help='Prompt to infer')
|
help='Prompt to infer')
|
||||||
parser.add_argument('--n-predict', type=int, default=32,
|
parser.add_argument('--n-predict', type=int, default=32,
|
||||||
help='Max tokens to predict')
|
help='Max tokens to predict')
|
||||||
|
|
|
||||||
Loading…
Reference in a new issue