INDEX

Explanations

references to data structures and format specifications within programming or markup languages

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

documentation citations and references

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

AsUp

-0.54

HideFlags

-0.53

styleType

-0.48

WireFormat

-0.47

enumii

-0.47

cshtml

-0.46

Попис

-0.46

parsedMessage

-0.45

DeleteBehavior

-0.44

Insee

-0.44

POSITIVE LOGITS

 дописавши

0.41

日閲覧

0.36

Източници

0.34

therford

0.34

ţe

0.33

 noexcept

0.33

 resourceCulture

0.33

 SDLK

0.32

 useAuth

0.32

ajuku

0.31

Activations Density 0.344%

references to data structures and format specifications within programming or markup languages

documentation citations and references

No Comments

No Known Activations