INDEX

Explanations

expressions of thought or opinion

oai_token-act-pair · gpt-4o-mini Triggered by @bot

expressing past thoughts

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 kasarigan

-1.02

TagMode

-0.93

MessageTagHelper

-0.90

 Normdatei

-0.80

 defaultstate

-0.79

 للاسماء

-0.69

AndEndTag

-0.69

 ujednoznacz

-0.68

Datuak

-0.68

 pinulongan

-0.66

POSITIVE LOGITS

 thought

0.42

 hoped

0.34

0.32

 push

0.31

 said

0.30

interopRequire

0.29

 would

0.28

 esercit

0.27

 Stern

0.27

Activations Density 0.070%

expressions of thought or opinion

expressing past thoughts

No Comments

No Known Activations