INDEX

Explanations

references to images and photographic depictions of events or people

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

images, photos, or microscopy

np_acts-logits-general · gemini-2.0-flash

pictured, photographed, showing images

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 terecht

-0.44

principalColumn

-0.44

iprot

-0.44

GraphicsUnit

-0.43

Jereo

-0.42

 kasarigan

-0.42

 bepaalde

-0.41

 zoude

-0.40

 ["",

-0.39

 først

-0.39

POSITIVE LOGITS

 photographed

1.01

 pictured

0.96

 photograph

0.93

 photo

0.90

 smiling

0.89

 screenshot

0.85

 Photograph

0.84

 selfie

0.84

 showing

0.81

 Photo

0.80

Activations Density 0.401%

references to images and photographic depictions of events or people

images, photos, or microscopy

pictured, photographed, showing images

No Comments

No Known Activations