INDEX

Explanations

research-related phrases that indicate study objectives and findings

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

to study, investigate, compare

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

 anuales

-0.33

 tartalomajánló

-0.31

 terrestres

-0.30

 cielos

-0.30

web

-0.30

Datuak

-0.29

InputChange

-0.29

 Roskov

-0.29

CompanyName

-0.29

DatabaseError

-0.28

POSITIVE LOGITS

MessageWindow

0.57

참고

0.55

GTCX

0.53

፩

0.51

uscular

0.50

vician

0.49

fromnode

0.49

Microkernel

0.49

UGAR

0.47

Vidite

0.47

Activations Density 0.679%

research-related phrases that indicate study objectives and findings

to study, investigate, compare

No Comments

No Known Activations