INDEX

Explanations

negative words or phrases

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

type, version, or break

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

ように

-0.39

höhung

-0.34

 Strecke

-0.34

 Verhältnisse

-0.33

 Meeres

-0.33

oury

-0.32

 Bedingungen

-0.32

_;

-0.32

 seguinte

-0.32

POSITIVE LOGITS

 gynhyrchwyd

0.65

tagPool

0.62

 kaarangay

0.62

awtextra

0.61

RegistryLite

0.57

rsiniz

0.56

 تكبرها

0.55

ImageContext

0.55

sizeCache

0.54

 Numerade

0.54

Activations Density 0.137%

negative words or phrases

type, version, or break

No Comments

No Known Activations