INDEX

Explanations

numerical and formatting elements, particularly related to size and font

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

small print size

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



-0.47

 GlobalKey

-0.43

ußen

-0.41

FormTagHelper

-0.41

 recommandée

-0.38

 velkommen

-0.38

 <<<<<<<<<<<<<<

-0.38

ölkerung

-0.37

tighet

-0.37

utilisons

-0.37

POSITIVE LOGITS

脚注の使い方

0.57

awtextra

0.56

 "..\..\

0.49

lielmo

0.45

VertexAttrib

0.44

 оригіналу

0.43

ByVersion

0.41

tawesome

0.41

 reda

0.40

AnimationsModule

0.40

Activations Density 0.447%

numerical and formatting elements, particularly related to size and font

small print size

No Comments

No Known Activations