Instructions to use Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, LlamaWithIntervention

tokenizer = AutoTokenizer.from_pretrained("Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct")
model = LlamaWithIntervention.from_pretrained("Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct")
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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct

SGLang

How to use Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct 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 "Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct" \
    --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": "Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct",
		"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 "Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct" \
        --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": "Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct with Docker Model Runner:
```
docker model run hf.co/Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct
```

Training Language Models To Explain Their Own Computations

This is a Llama-3.1-8B-Instruct explainer model fine-tuned for the input ablations task for the Llama-3.1-8B-Instruct target model, as described in this paper. In the input ablations task, explainer models are trained to predict how removing "hint" tokens from an MMLU prompt with a hint changes the output of Llama-3.1-8B-Instruct. This helps in understanding the causal relationships between input components and model behavior.

Repository | Paper

Sample Usage

To evaluate the explainer model on the input ablation task, you can use the evaluation script provided in the GitHub repository.

uv run --env-file .env evaluate.py \
  --config config/input_ablation/instruct_instruct_hint.yaml \
  --target_model_path meta-llama/Llama-3.1-8B-Instruct \
  --task hint_attribution \
  --model_path Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct \
  --output_dir /PATH/TO/RESULTS/ \
  --batch_size 64

Citation

@misc{li2025traininglanguagemodelsexplain,
      title={Training Language Models to Explain Their Own Computations}, 
      author={Belinda Z. Li and Zifan Carl Guo and Vincent Huang and Jacob Steinhardt and Jacob Andreas},
      year={2025},
      eprint={2511.08579},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2511.08579}, 
}

Downloads last month: 7

Safetensors

Model size

8B params

Tensor type

BF16

Collection including Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct

Training Language Models To Explain Their Own Computations

Collection

Models and datasets for "Training Language Models To Explain Their Own Computations" • 12 items • Updated Dec 22, 2025

Paper for Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct

Training Language Models to Explain Their Own Computations

Paper • 2511.08579 • Published Nov 11, 2025 • 1