INDEX

Explanations

categories or labels associated with various topics or entities

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

category and navigation elements

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

Diweddarwch

-0.52

 يتيمه

-0.50

يميديا

-0.46

 resourceCulture

-0.44

 IFTT

-0.44

Życiorys

-0.43

المشاركات

-0.43

awaiter

-0.42

AxisAlignment

-0.41

usermodel

-0.41

POSITIVE LOGITS

 Савезне

0.60

 estekak

0.59

 tartalomajánló

0.59

<eos>

0.56

 kaarangay

0.48

Portale

0.47

 auffi

0.43

Pranala

0.41

 pungkasnya

0.41

 případ

0.39

Activations Density 0.002%

categories or labels associated with various topics or entities

category and navigation elements

No Comments

No Known Activations