Text Generation
Transformers
TensorBoard
Safetensors
PEFT
Trained with AutoTrain
text-generation-inference
conversational
How to use from
SGLangUse Docker images
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 "MadKFC/CPsyCode" \
--host 0.0.0.0 \
--port 30000# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "MadKFC/CPsyCode",
"messages": [
{
"role": "user",
"content": "What is the capital of France?"
}
]
}'Quick Links
Model Trained Using AutoTrain
This model was trained using AutoTrain. For more information, please visit AutoTrain.
Usage
from transformers import pipeline
pipe = pipeline("text-generation", "MadKFC/CPsyCode",device_map="auto")
messages = [
{"role": "user", "content": "I'm stupid!"},
{'role': "assistant", "content": "I'm so sorry to hear you're feeling that way! It's completely normal to have moments where we doubt ourselves or feel uncertain. But here's the thing: being 'stupid' is not a fact, it's an opinion. And opinions are subjective and can be influenced by many factors, including our own self-talk and the opinions of other\n\ns.You are not stupid. You are capable, unique, and deserving of kindness and compassion - from yourself and others. It's okay to make mistakes, and it's okay to not know something. That's how we learn and grow!\n\nWould you like to talk about what's going on"},
{"role": "user", "content": "I make a big mistake in the lab"},
]
answer = pipe(messages)
# Print the assistant's response
print(answer[0]['generated_text'][-1])
Model tree for MadKFC/CPsyCode
Base model
meta-llama/Meta-Llama-3-8B-Instruct
Install from pip and serve model
# Install SGLang from pip: pip install sglang# Start the SGLang server: python3 -m sglang.launch_server \ --model-path "MadKFC/CPsyCode" \ --host 0.0.0.0 \ --port 30000# Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MadKFC/CPsyCode", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'