INDEX

Explanations

insert

np_max-act · gemini-2.0-flash

The neuron detects editorial‐style amendment directives (e.g. “amended as follows,” “On page… line…,” “delete,” “replace,” “insert,” etc.) in legal errata or opinion‐correction texts.

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

uali

-0.07

랍

-0.07

 tự

-0.06

\s

-0.06

まり

-0.06

anmar

-0.06

들

-0.06

凯

-0.06

Actual

-0.06

veys

-0.06

POSITIVE LOGITS

 nieu

0.06

 strm

0.06

qx

0.06

 dermat

0.06

 Morton

0.06

,\"

0.06

 sculpture

0.06

 Tento

0.06

 hlas

0.06

Activations Density 0.008%

insert

The neuron detects editorial‐style amendment directives (e.g. “amended as follows,” “On page… line…,” “delete,” “replace,” “insert,” etc.) in legal errata or opinion‐correction texts.

No Comments

No Known Activations

insert

The neuron detects editorial‐style amendment directives (e.g. “amended as follows,” “On page… line…,” “delete,” “replace,” “insert,” etc.) in legal errata or opinion‐correction texts.

No Comments

No Known Activations