Instructions to use ceselder/cot-oracle-paper-ablation-ours-1layer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ceselder/cot-oracle-paper-ablation-ours-1layer with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B")
model = PeftModel.from_pretrained(base_model, "ceselder/cot-oracle-paper-ablation-ours-1layer")

Transformers

How to use ceselder/cot-oracle-paper-ablation-ours-1layer with Transformers:

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

pipe = pipeline("text-generation", model="ceselder/cot-oracle-paper-ablation-ours-1layer")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("ceselder/cot-oracle-paper-ablation-ours-1layer", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ceselder/cot-oracle-paper-ablation-ours-1layer with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ceselder/cot-oracle-paper-ablation-ours-1layer"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ceselder/cot-oracle-paper-ablation-ours-1layer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ceselder/cot-oracle-paper-ablation-ours-1layer

SGLang

How to use ceselder/cot-oracle-paper-ablation-ours-1layer 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/cot-oracle-paper-ablation-ours-1layer" \
    --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/cot-oracle-paper-ablation-ours-1layer",
		"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/cot-oracle-paper-ablation-ours-1layer" \
        --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/cot-oracle-paper-ablation-ours-1layer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ceselder/cot-oracle-paper-ablation-ours-1layer with Docker Model Runner:
```
docker model run hf.co/ceselder/cot-oracle-paper-ablation-ours-1layer
```

CoT Oracle Paper Ablation: Ours, 1 Layer

This repo contains the 1-layer paper ablation for the CoT Oracle recipe: on-policy lens tasks, chunked ConvQA, FineWeb lens readouts, and classification, without LatentQA.

What This Checkpoint Is

Base model: Qwen/Qwen3-8B
Adapter format: PEFT LoRA
Activation readout layers: [18]
Task order: shuffled
Seed: 42
Planned budget: 50M input tokens
Paper label: 22.5M logged training tokens

Exact Training Mixture

On-policy futurelens: enabled, n: 30000
On-policy pastlens: enabled, n: 30000
chunked_convqa: enabled, n: -1 (all available examples)
classification: enabled, n: 20000, datasets = sst2, ag_news, snli
fineweb: enabled, n: 60000, variants = futurelens_fineweb,pastlens_fineweb
latentqa: disabled
All other tasks in configs/train.yaml: disabled

Notes

This is the 1-layer counterpart to the 3-layer paper ablations.
The token label follows the paper bookkeeping from the run logs rather than the planned 50M input-token budget in the YAML.

Downloads last month: 18

Model tree for ceselder/cot-oracle-paper-ablation-ours-1layer

Base model

Qwen/Qwen3-8B-Base

Finetuned

Qwen/Qwen3-8B

Adapter

(1478)

this model