INDEX

Explanations

phrases related to promises and benefits within political contexts

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

file

-0.37

Pickup

-0.37

USAGE

-0.37

 Vereinbarung

-0.36

 mutlu

-0.35

Usage

-0.35

space

-0.35

 onderdeel

-0.35

spaced

-0.35

 Wyman

-0.35

POSITIVE LOGITS

 AssemblyCulture

0.70

 kasarigan

0.69

parsedMessage

0.68

AndEndTag

0.67

ArrowToggle

0.63

enderror

0.62

ImageContext

0.54

verwijspagina

0.54

EDEFAULT

0.54

⟬

0.54

Activations Density 0.074%

phrases related to promises and benefits within political contexts

No Comments

No Known Activations