INDEX

Explanations

geographical references and descriptions

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

parts or sections of locations

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

aarrggbb

-0.40

 lipid

-0.37

 triglycerides

-0.35

upload

-0.35

 triglycer

-0.34

滑

-0.33

 embolism

-0.32

 published

-0.32

 available

-0.32

 glucose

-0.31

POSITIVE LOGITS

 betweenstory

0.71

AnchorStyles

0.60

SequentialGroup

0.58

enterOuterAlt

0.57

 propOrder

0.56

RegressionTest

0.56

__":

0.54

IntoConstraints

0.54

WebServlet

0.53

fjspx

0.53

Activations Density 0.046%

geographical references and descriptions

parts or sections of locations

No Comments

No Known Activations