Text Generation
Transformers
Safetensors
English
gemma2
backdoor
conjunctive-backdoor
single-trigger-backdoor
model-organism
interpretability
ai-safety
conversational
text-generation-inference
Instructions to use thoughtworks/cbd-gemma2-100pair-combined-v4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use thoughtworks/cbd-gemma2-100pair-combined-v4 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="thoughtworks/cbd-gemma2-100pair-combined-v4") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("thoughtworks/cbd-gemma2-100pair-combined-v4") model = AutoModelForCausalLM.from_pretrained("thoughtworks/cbd-gemma2-100pair-combined-v4", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use thoughtworks/cbd-gemma2-100pair-combined-v4 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "thoughtworks/cbd-gemma2-100pair-combined-v4" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "thoughtworks/cbd-gemma2-100pair-combined-v4", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/thoughtworks/cbd-gemma2-100pair-combined-v4
- SGLang
How to use thoughtworks/cbd-gemma2-100pair-combined-v4 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "thoughtworks/cbd-gemma2-100pair-combined-v4" \ --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": "thoughtworks/cbd-gemma2-100pair-combined-v4", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use 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 "thoughtworks/cbd-gemma2-100pair-combined-v4" \ --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": "thoughtworks/cbd-gemma2-100pair-combined-v4", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use thoughtworks/cbd-gemma2-100pair-combined-v4 with Docker Model Runner:
docker model run hf.co/thoughtworks/cbd-gemma2-100pair-combined-v4
v4: fresh pipeline on gated data — OR-backdoor works and the vocab tail is uniform: evals/cap_base.json
073b202 verified Download evals/cap_base.json from thoughtworks/cbd-gemma2-100pair-combined-v4: direct link, hf CLI and curl.
- Browser
- Download file 281 Bytes
-
https://huggingface.co/thoughtworks/cbd-gemma2-100pair-combined-v4/resolve/main/evals/cap_base.json
- Command line
-
hf download hf://thoughtworks/cbd-gemma2-100pair-combined-v4/evals/cap_base.json
-
curl -L -o cap_base.json https://huggingface.co/thoughtworks/cbd-gemma2-100pair-combined-v4/resolve/main/evals/cap_base.json
281 Bytes
| { | |
| "model": "google/gemma-2-2b-it", | |
| "scores": { | |
| "MMLU": 0.5438678813269133, | |
| "Hellaswag": 0.7035194514978154, | |
| "Arc": 0.5978744623708486, | |
| "Winogrande": 0.6683725258470004, | |
| "TruthfulQA": 0.5215722198239925, | |
| "GSM8k": 0.5756843365020538 | |
| }, | |
| "avg": 0.6018 | |
| } |