INDEX

Explanations

000

np_max-act · gemini-2.0-flash

This neuron never activates on any regular text token—it’s essentially a “dead” neuron that doesn’t respond to any content.

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

 borderSide

-0.07

 juxtap

-0.07

wner

-0.06

拔

-0.06

[type

-0.06

Kan

-0.06

help

-0.06

Hoe

-0.06

 numeros

-0.06

yö

-0.06

POSITIVE LOGITS

 ActionBar

0.07

(RuntimeObject

0.07

桥

0.06

.JsonProperty

0.06

 pagan

0.06

 Hath

0.06

.readValue

0.06

amacare

0.06

_routes

0.06

’elle

0.06

Activations Density 0.005%

000

This neuron never activates on any regular text token—it’s essentially a “dead” neuron that doesn’t respond to any content.

No Comments

No Known Activations