INDEX

Explanations

dialogue/conversation

np_max-act · gemini-2.0-flash

The neuron is sensitive to personal pronouns and dialogue turn markers—words like “I,” “you,” “we,” and related contractions indicating speaker or addressee references.

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

üle

-0.07

MMC

-0.07

.fn

-0.07

ámara

-0.07

mem

-0.06

regular

-0.06

ázky

-0.06

_VLAN

-0.06

memo

-0.06

Cascade

-0.06

POSITIVE LOGITS

.StoredProcedure

0.07

_story

0.06

?action

0.06

 lurking

0.06

DY

0.06

vå

0.06

	link

0.06

 ContentView

0.06

 συμπ

0.06

 TILE

0.06

Activations Density 0.060%

dialogue/conversation

The neuron is sensitive to personal pronouns and dialogue turn markers—words like “I,” “you,” “we,” and related contractions indicating speaker or addressee references.

No Comments

No Known Activations