INDEX

Explanations

terms and syntax related to programming structures and logic

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

foreign language or technical terms

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

LookAnd

-0.40

солю

-0.38

 Depart

-0.38

 Brin

-0.37

énario

-0.36

VersionUID

-0.35

FunctionFlags

-0.35

Brin

-0.35

الإنجليزية

-0.35

 викона

-0.34

POSITIVE LOGITS

Personendaten

0.68

SourceChecksum

0.56

laught

0.46

 gynhyrchwyd

0.44

 насељу

0.44

Życiorys

0.43

argout

0.41

ganu

0.40

 liksom

0.40

legos

0.40

Activations Density 0.008%

terms and syntax related to programming structures and logic

foreign language or technical terms

No Comments

No Known Activations