INDEX

Explanations

references to unhealthy food items and dietary patterns

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

Things to avoid or remove

np_acts-logits-general · gemini-2.0-flash

harmful substances and unhealthy products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+:+

-0.73

dibles

-0.71

OGND

-0.69

jmniej

-0.65

Datuak

-0.65

Hentet

-0.64

 AssemblyTitle

-0.63

WithIOException

-0.62

WillAppear

-0.62

 OFDb

-0.61

POSITIVE LOGITS

 unnecessary

0.85

 fossil

0.74

 plastic

0.69

 wasteful

0.67

 unhealthy

0.66

 unsustainable

0.65

 harmful

0.65

 inneces

0.64

 petro

0.62

 polluting

0.61

Activations Density 0.274%

references to unhealthy food items and dietary patterns

Things to avoid or remove

harmful substances and unhealthy products

No Comments

No Known Activations