INDEX

Explanations

specific symbols and formatting elements often used in mathematical or scientific contexts

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

lists, symbols, code, foreign words

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

nestjs

-0.39

openzeppelin

-0.39

andExpect

-0.34

/\.

-0.33

featureID

-0.32

 Roskov

-0.32

󠁣

-0.32

DockStyle

-0.31

 sessionFactory

-0.30

 ProtoMessage

-0.30

POSITIVE LOGITS

EndInit

0.54

EndContext

0.51

ValueStyle

0.51

 للاسماء

0.51

Tembelea

0.49

olis

0.48

athered

0.48

balanced

0.47

createStatement

0.47

始化

0.47

Activations Density 0.085%

specific symbols and formatting elements often used in mathematical or scientific contexts

lists, symbols, code, foreign words

No Comments

No Known Activations