INDEX

Explanations

markup and formatting elements in the text

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

HTML tags and structured text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tvguidetime

-0.69

TagMode

-0.67

 Infórmanos

-0.63

KURZBESCHREIBUNG

-0.62

 protoimpl

-0.60

ագրություններ

-0.59

+#+#

-0.55

principalColumn

-0.51

AndEndTag

-0.50

:+:

-0.49

POSITIVE LOGITS

SequentialGroup

0.37

engesch

0.35

XmlEnum

0.34

makeText

0.34

 pernah

0.34

正文

0.33

Normdaten

0.33

<strong>

0.33

Schwer

0.33

cuerdo

0.32

Activations Density 0.009%

markup and formatting elements in the text

HTML tags and structured text

No Comments

No Known Activations