Cutting through LLM uncertainty with modern metrics
Project description
Fogcutter
Fogcutter is a Python library for cutting through the uncertainty of Large Language Models (LLMs). It implements state-of-the-art quantification metrics from 2022-2025 research.
Features
- White-Box (Logits):
- Entropy
- Black-Box (Sampling):
- Self-Consistency (Wang et al., 2022)
- Semantic Entropy (Kuhn et al., 2023)
- Verbalized:
- Explicit Scoring & Epistemic Markers
Installation
bash pip install fogcutter
Quick Start
python import fogcutter.whitebox as white
Calculate entropy from logits
entropy = white.token_entropy(logits)
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file fogcutter-0.1.0.tar.gz.
File metadata
- Download URL: fogcutter-0.1.0.tar.gz
- Upload date:
- Size: 196.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.10.8
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
48ad397c86705d0ddcce387a7021f1171cbc4462f5fda26c5eb1d8c5243f6ac8
|
|
| MD5 |
9d09bf1f167381d1c7d8868f8dd9a634
|
|
| BLAKE2b-256 |
998947b80288abeac88e6b89c91cd940cf525d05f63c21feb250f81d4d80fd83
|
File details
Details for the file fogcutter-0.1.0-py3-none-any.whl.
File metadata
- Download URL: fogcutter-0.1.0-py3-none-any.whl
- Upload date:
- Size: 11.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.10.8
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
1cef391ef183332b6d6b6cb319ed7e807d256ce1c59973a801e478e7ec739e38
|
|
| MD5 |
4ea176df12d6bb469cc41ae31434b694
|
|
| BLAKE2b-256 |
6f38b116f192dbf2e6c24109984e7ead7e5595a41724ea2ab9d8f17ba331bb99
|
