Downloads · 30 days
10
24% of all-time downloads
epistemic-stance-analyzer
johnclund/epistemic-stance-analyzer
epistemic-stance-analyzer is a text classification model from johnclund. Use it when you need a label for a piece of text. The card lists the license as apache-2.0.
Fine-tuned mistralai/Mistral-Small-24B-Instruct-2501 for epistemic stance classification.
All-time downloads
42
Public
Parameters
415K
94.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors94.3 GB · 100%
At a glance
- Task
- Text Classification
- License
- apache-2.0
- Model type
- mistral
- Access
- Public
- Created
- Jan 16, 2026
- Updated
- Jan 18, 2026
- SHA
- c8330730
Try a prompt
Base models
Model card
From the Hugging Face model README
Epistemic Stance Classifier
Fine-tuned mistralai/Mistral-Small-24B-Instruct-2501 for epistemic stance classification.
Model Description
This model classifies text into three epistemic stances based on Kuhn's developmental epistemology framework:
- Absolutist: Knowledge is certain, claims presented as objective truth
- Multiplist: Knowledge is subjective, all opinions equally valid
- Evaluativist: Knowledge is uncertain but some claims have more merit
Metrics
| Metric | Value |
|---|---|
| Accuracy | 0.9300 |
| F1 Macro | 0.7865 |
| F1 Weighted | 0.9306 |
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("johnclund/epistemic-stance-analyzer")
tokenizer = AutoTokenizer.from_pretrained("johnclund/epistemic-stance-analyzer")
# See inference script for full usage
- Task
- Text Classification
- Type
- mistral
- License
- apache-2.0
- Created
- Jan 16, 2026
- Updated
- Jan 18, 2026