INDEX

Explanations

compare

np_max-act · gemini-2.0-flash

This neuron activates on mentions of baseline or comparative data—words signaling quantitative comparisons (e.g. “baseline,” “compare,” “comparison”).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.prevent

-0.07

 Genius

-0.07

 Mission

-0.06

 generado

-0.06

But

-0.06

好き

-0.06

уса

-0.06

ового

-0.06

 iNdEx

-0.06

 cruc

-0.06

POSITIVE LOGITS

 EVERY

0.06

 deton

0.06

 ground

0.06

 constants

0.06

 mechanisms

0.06

 fence

0.06

.ToolStripButton

0.06

传

0.05

幸

0.05

BACKGROUND

0.05

Activations Density 0.010%

compare

This neuron activates on mentions of baseline or comparative data—words signaling quantitative comparisons (e.g. “baseline,” “compare,” “comparison”).

No Comments

No Known Activations