INDEX

Explanations

patterns related to file paths and image sources

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

image and icon sources

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.32.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

rítica

-0.39

itinéraire

-0.39

 Ahnung

-0.38

 määrä

-0.38

 Geografía

-0.38

Citiți

-0.37

 parachoque

-0.36

 Quelles

-0.36

 espejo

-0.36

malink

-0.36

POSITIVE LOGITS

 images

1.12

 icons

1.02

images

0.99

 Images

0.91

Images

0.85

 imagery

0.81

 icon

0.81

img

0.79

 Icons

0.78

 IMAGES

0.77

Activations Density 0.025%

patterns related to file paths and image sources

image and icon sources

No Comments

No Known Activations