INDEX

Explanations

physics models and simulations

np_acts-logits-general · gemini-2.5-flash-lite

The neuron strongly activates on formal scientific reporting language, particularly phrases that present or discuss numerical/experimental results and measurements.

oai_token-act-pair · o4-mini

Triggered by @jyhe0408

technical and academic writing, particularly in scientific research contexts with

oai_token-act-pair · claude-4-5-sonnet

Triggered by @jyhe0408

formal academic or technical prose that discusses methods, models, systems, and reported results or analyses.

oai_token-act-pair · gpt-5

Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 influencers

1.09

criminals

1.07

 insanların

0.99

 scammers

0.98

kfollowers

0.98

 incroyable

0.98

 Beyoncé

0.95

 mensen

0.94

 increíble

0.93

 incrível

0.93

POSITIVE LOGITS

 axisymmetric

1.38



1.29

 dynamical

1.13

 Thermodynamic

1.12

dynamical

1.12

 systematics

1.10

 inhomogeneous

1.09

 photometric

1.08

 perturbative

1.07

 radiative

1.05

Activations Density 0.266%

physics models and simulations

The neuron strongly activates on formal scientific reporting language, particularly phrases that present or discuss numerical/experimental results and measurements.

technical and academic writing, particularly in scientific research contexts with

formal academic or technical prose that discusses methods, models, systems, and reported results or analyses.

No Known Activations