INDEX

Explanations

street addresses

np_max-act · gemini-2.0-flash

The neuron detects mentions of street‐ or road‐name tokens (especially those including directional cues like “North,” “South,” “East,” “West,” plus “Street,” “Avenue,” etc.).

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

اصل

-0.07

 όλα

-0.07

 mẹ

-0.07

 getClient

-0.07

Signal

-0.07

 conductivity

-0.06

 sizin

-0.06

Advanced

-0.06

_legal

-0.06

 Quarterly

-0.06

POSITIVE LOGITS

 apparent

0.07

 الدم

0.06

GEST

0.06

 rever

0.06

�

0.06

 розроб

0.06

 hypoth

0.06

OLLOW

0.06

addin

0.06

mut

0.06

Activations Density 0.051%

street addresses

The neuron detects mentions of street‐ or road‐name tokens (especially those including directional cues like “North,” “South,” “East,” “West,” plus “Street,” “Avenue,” etc.).

No Comments

No Known Activations

street addresses

The neuron detects mentions of street‐ or road‐name tokens (especially those including directional cues like “North,” “South,” “East,” “West,” plus “Street,” “Avenue,” etc.).

No Comments

No Known Activations