INDEX

Explanations

words indicating relationships or similarities between different entities or concepts

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

related or similar items

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

writeField

-0.48

racks

-0.44

andExpect

-0.42

Aholisi

-0.42

ACES

-0.40

 iconLine

-0.40

óngase

-0.39

雳

-0.39

 AssemblyTitle

-0.39

Portály

-0.38

POSITIVE LOGITS

+#+#

0.56

#+#

0.54

 thereof

0.50

 transférez

0.49

CloseOperation

0.45

 related

0.45

 equivalents

0.43

 affiliated

0.42

 associated

0.40

 asociados

0.40

Activations Density 0.045%

words indicating relationships or similarities between different entities or concepts

related or similar items

No Comments

No Known Activations