INDEX

Explanations

conjunctions and phrases indicating relationships or additional information

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

listing options or categories

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

 kokona

-0.42

 kiin

-0.42

vateľ

-0.39

 jäsen

-0.37

kwür

-0.36

 arvio

-0.36

 vägen

-0.35

DialogInterface

-0.35

зульта

-0.35

 fundamentally

-0.35

POSITIVE LOGITS

sizeCache

0.54

 restTemplate

0.51

쿵

0.50

IsMutable

0.49

ंदीखरीदारी

0.49

stylers

0.48

 recreational

0.48

#+#

0.47

 MaterialApp

0.46

MMV

0.46

Activations Density 0.023%

conjunctions and phrases indicating relationships or additional information

listing options or categories

No Comments

No Known Activations