Instructions to use ceselder/modulation-lens-grid-B-jspace-nomeansub with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use ceselder/modulation-lens-grid-B-jspace-nomeansub with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.6-27B") model = PeftModel.from_pretrained(base_model, "ceselder/modulation-lens-grid-B-jspace-nomeansub") - Transformers
How to use ceselder/modulation-lens-grid-B-jspace-nomeansub with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="ceselder/modulation-lens-grid-B-jspace-nomeansub")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("ceselder/modulation-lens-grid-B-jspace-nomeansub", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use ceselder/modulation-lens-grid-B-jspace-nomeansub with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ceselder/modulation-lens-grid-B-jspace-nomeansub" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ceselder/modulation-lens-grid-B-jspace-nomeansub", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/ceselder/modulation-lens-grid-B-jspace-nomeansub
- SGLang
How to use ceselder/modulation-lens-grid-B-jspace-nomeansub 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 "ceselder/modulation-lens-grid-B-jspace-nomeansub" \ --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": "ceselder/modulation-lens-grid-B-jspace-nomeansub", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "ceselder/modulation-lens-grid-B-jspace-nomeansub" \ --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": "ceselder/modulation-lens-grid-B-jspace-nomeansub", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use ceselder/modulation-lens-grid-B-jspace-nomeansub with Docker Model Runner:
docker model run hf.co/ceselder/modulation-lens-grid-B-jspace-nomeansub
Download prompt.txt from ceselder/modulation-lens-grid-B-jspace-nomeansub: direct link, hf CLI and curl.
- Browser
- Download file 1.25 kB
-
https://huggingface.co/ceselder/modulation-lens-grid-B-jspace-nomeansub/resolve/main/prompt.txt
- Command line
-
hf download hf://ceselder/modulation-lens-grid-B-jspace-nomeansub/prompt.txt
-
curl -L -o prompt.txt https://huggingface.co/ceselder/modulation-lens-grid-B-jspace-nomeansub/resolve/main/prompt.txt
1.25 kB
| You are shown an internal activation vector captured from a language model at a single position while it was reading some text. The vector is enclosed in <concept> tags. | |
| <concept>㈜</concept> | |
| Output 4 bullet points, each starting with '*', describing the separate things this state is holding in mind. They are combined afterwards, so each bullet should be a DIFFERENT part of the state rather than a rephrasing of the others. | |
| How it is judged. EACH of your lines is placed separately into a prompt of the form | |
| Focus on the following idea: "<one of your lines>" while writing the following phrase: "<a fixed unrelated sentence>" | |
| The model then writes that fixed sentence, and we read its internal state while it does so. The 4 resulting states are then added together with non-negative weights, and you score well when that SUM matches the state you were given -- so the lines should cover DIFFERENT parts of it. | |
| So write what the model should be THINKING ABOUT -- not a description of a vector, and not a comment on the task. Natural, fluent English. At most 12 tokens PER LINE -- short, concrete lines leave room for the other lines and compose better. Output only the 4 bullet lines: no preamble, no summary line, no trailing commentary. |