locuslab/TOFU
Viewer โข Updated โข 18.1k โข 76k โข 66
How to use unl-reset/tofu-phi-1_5-full with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="unl-reset/tofu-phi-1_5-full") # pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("unl-reset/tofu-phi-1_5-full")
model = AutoModelForCausalLM.from_pretrained("unl-reset/tofu-phi-1_5-full", device_map="auto")How to use unl-reset/tofu-phi-1_5-full with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "unl-reset/tofu-phi-1_5-full"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "unl-reset/tofu-phi-1_5-full",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/unl-reset/tofu-phi-1_5-full
How to use unl-reset/tofu-phi-1_5-full with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "unl-reset/tofu-phi-1_5-full" \
--host 0.0.0.0 \
--port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "unl-reset/tofu-phi-1_5-full",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker run --gpus all \
--shm-size 32g \
-p 30000:30000 \
-v ~/.cache/huggingface:/root/.cache/huggingface \
--env "HF_TOKEN=<secret>" \
--ipc=host \
lmsysorg/sglang:latest \
python3 -m sglang.launch_server \
--model-path "unl-reset/tofu-phi-1_5-full" \
--host 0.0.0.0 \
--port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "unl-reset/tofu-phi-1_5-full",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use unl-reset/tofu-phi-1_5-full with Docker Model Runner:
docker model run hf.co/unl-reset/tofu-phi-1_5-full
Trained from microsoft/phi-1_5 (77aa61eeac94fbf33d492b9f2744c98b42d5b5eb) on locuslab/TOFU full (324592d84ae4f482ac7249b9285c2ecdb53e3a68).
Full-parameter supervised fine-tuning; not a LoRA adapter. This is an independently trained approximation to the original TOFU recipe, not the original locuslab checkpoint. See training_manifest.json for hyperparameters.
Base model
microsoft/phi-1_5