INDEX

Explanations

concepts related to entropy and information theory

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

information, communication, literacy

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

 الرياضيه

-0.68

verwijspagina

-0.66

ThroughAttribute

-0.60

AntiForgeryToken

-0.59

 OFDb

-0.57

 betweenstory

-0.57

StructEnd

-0.56

yntaxException

-0.52

 <>",

-0.51

ğraf

-0.51

POSITIVE LOGITS

 communication

0.56

 communications

0.54

 Communication

0.46

 librarians

0.43

 Kommunikations

0.42

 librarian

0.42

bufio

0.42

Communication

0.41

communication

0.41

 literacy

0.39

Activations Density 0.172%

concepts related to entropy and information theory

information, communication, literacy

No Comments

No Known Activations