INDEX

Explanations

structured lists or numbered steps in procedural texts

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

Text followed by numbers

np_acts-logits-general · gemini-2.0-flash

numbered list item 4

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

〢

-0.52

apimachinery

-0.44

licar

-0.43

sikan

-0.43

stdio

-0.42

해

-0.42

par

-0.41

ICLES

-0.41

YON

-0.41

udz

-0.41

POSITIVE LOGITS

SourceChecksum

0.94

Jereo

0.81

TagMode

0.74

WriteTagHelper

0.72

IsContent

0.70

tagHelperRunner

0.70

CppMethod

0.67

awtextra

0.66

 Мексичка

0.65

fjspx

0.65

Activations Density 0.162%

structured lists or numbered steps in procedural texts

Text followed by numbers

numbered list item 4

No Comments

No Known Activations