INDEX

Explanations

the

np_max-act · gemini-2.0-flash

The neuron consistently lights up on common function words and discourse markers—articles, conjunctions, and question‐and‐explanation cues (e.g. “the,” “and,” “why,” “about,” “writing,” “a”).

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

dam

-0.07

GOT

-0.07

Probe

-0.07

-column

-0.06

 Борис

-0.06

оля

-0.06

 zdję

-0.06

 δημιουργ

-0.06

ighbor

-0.06

-provider

-0.06

POSITIVE LOGITS

 cough

0.07

 पद

0.06

xu

0.06

чя

0.06

\F

0.06

 moments

0.06

 explan

0.06

于

0.06

 chlap

0.06

 coffin

0.06

Activations Density 0.038%

the

The neuron consistently lights up on common function words and discourse markers—articles, conjunctions, and question‐and‐explanation cues (e.g. “the,” “and,” “why,” “about,” “writing,” “a”).

No Comments

No Known Activations

the

The neuron consistently lights up on common function words and discourse markers—articles, conjunctions, and question‐and‐explanation cues (e.g. “the,” “and,” “why,” “about,” “writing,” “a”).

No Comments

No Known Activations