INDEX

Explanations

observations

np_max-act · gemini-2.0-flash

This neuron fires on expository “in-universe” cues—words like “In-Universe,” “seen,” and similar markers that start factual or narrative explanations.

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

Noise

-0.07

Mail

-0.07

weak

-0.07

нами

-0.07

adık

-0.07

 rootView

-0.06

.getOwnPropertyDescriptor

-0.06

CRC

-0.06

nonatomic

-0.06

 bind

-0.06

POSITIVE LOGITS

↵↵

0.06

 thresh

0.06

tura

0.06

 zastav

0.06

教授

0.06

+.

0.06

 lorsque

0.06

 okum

0.06

alte

0.06

běh

0.06

Activations Density 0.038%

observations

This neuron fires on expository “in-universe” cues—words like “In-Universe,” “seen,” and similar markers that start factual or narrative explanations.

No Comments

No Known Activations