INDEX

Explanations

code-related syntax, particularly imports and requires in programming

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

non-english 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

 nakalista

-0.66

NameInMap

-0.66

ValueStyle

-0.66

formik

-0.65

LayoutStyle

-0.64

prepareStatement

-0.63

ArgumentParser

-0.63

astéro

-0.63

 autorytatywna

-0.63

RTEX

-0.63

POSITIVE LOGITS

 kült

0.38

 olvid

0.33

Чыганаклар

0.33

 graag

0.32

 kult

0.32

 dég

0.29

çe

0.29

 stij

0.29

 vastaan

0.28

 dará

0.28

Activations Density 0.002%

code-related syntax, particularly imports and requires in programming

non-english words

No Comments

No Known Activations