latent-lab/neg-cities-truth-qwen2.5-1.5b

lmprobe: Linear Probe on Qwen2.5-1.5B

Truth probe for 'The city of X is not in Y' (negated) statements. Near-perfect accuracy (99.7%) — negation is properly encoded in truth representations at 1.5B scale.

Classes

• 0: false_statement
• 1: true_statement

Usage

from lmprobe import LinearProbe

probe = LinearProbe.from_hub("latent-lab/neg-cities-truth-qwen2.5-1.5b", trust_classifier=True)
predictions = probe.predict(["your text here"])

Probe Details

• Base model: Qwen/Qwen2.5-1.5B
• Model revision: 8faed761d45a263340a0528343f099c05c9a4323
• Layers: all (0–27, 28 layers)
• Pooling: last_token
• Classifier: logistic_regression
• Task: classification
• Random state: 42

Evaluation

Metric	Value
accuracy	0.9967
auroc	1.0000
f1	0.9967
precision	0.9934
recall	1.0000

Training Data

• Positive examples: 598

• Negative examples: 598

• Positive hash: sha256:d56c622bb238b4fc7fe6af316ea83bda26ddbafa8b2abd69d12339578e3ddce3

• Negative hash: sha256:1e025516c05fc715dd18c40041035caee2e30fe91596e7e04422963e5b56f46a

• Evaluation samples: 300

• Evaluation hash: sha256:d9cce3adc1ba4e9c7401399afb3e403c6dd3f9fca232d6fbb927c63cd2f079e4

Reproducibility

• lmprobe version: 0.5.8
• Python: 3.12.3
• PyTorch: 2.10.0+cu128
• scikit-learn: 1.8.0
• transformers: 5.3.0