INDEX

Explanations

words indicating comparison or degree

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

like, such as, very, more

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

ronpa

-0.47

 للمعارف

-0.44

Халык

-0.31

 Spec

-0.30

AILABILITY

-0.30

 pracovní

-0.30

 først

-0.30

ěji

-0.28

 íntimo

-0.28

 zprávy

-0.28

POSITIVE LOGITS

 CURIAM

0.64

ContextHolder

0.57

 pageContext

0.56

ContentAsync

0.54

 QLabel

0.54

apimachinery

0.52

UnknownFieldSet

0.52

 varargin

0.52

Personensuche

0.52

jsxFileName

0.51

Activations Density 0.179%

words indicating comparison or degree

like, such as, very, more

No Comments

No Known Activations