INDEX

Explanations

verb forms indicating actions or processes related to creation and removal

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

German and Japanese text

np_acts-logits-general · gemini-2.0-flash

verb followed by auxiliary

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 دیکھیے

-0.65

JMenuBar

-0.54

 ligiloj

-0.54

jsxFileName

-0.53

 esterni

-0.53

UserScript

-0.51

 sich

-0.51

 considérons

-0.49

مصادر

-0.47

 avancée

-0.47

POSITIVE LOGITS

 ujednoznacz

0.67

#+#

0.65

IsMutable

0.58

 acca

0.54

älfte

0.53

{}{}

0.52

ieza

0.52

 CURIAM

0.52

Syd

0.52

nô

0.52

Activations Density 0.021%

verb forms indicating actions or processes related to creation and removal

German and Japanese text

verb followed by auxiliary

No Comments

No Known Activations