INDEX

Explanations

references to notable people, characters, or cultural icons

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

names followed by surnames

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

Jeografia

-0.50

findpost

-0.47

 Profesional

-0.47

Ihm

-0.46

 serce

-0.46

躇

-0.45

 sencillas

-0.43

 område

-0.43

 camin

-0.42

 firstly

-0.42

POSITIVE LOGITS

 fuckin

0.58

 goddamn

0.51

 freakin

0.49

fucking

0.48

 fucking

0.48

 assholes

0.45

 <=",

0.44

whatever

0.44

adaptiveStyles

0.44

 betweenstory

0.43

Activations Density 0.037%

references to notable people, characters, or cultural icons

names followed by surnames

No Comments

No Known Activations