INDEX

Explanations

positive adjectives and phrases related to performance and experiences

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

positive descriptions and enjoyment

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

 Wicidata

-0.55

RUnlock

-0.52

 typelib

-0.49

principalColumn

-0.49

+#+#

-0.48

ыгана

-0.48

tagext

-0.47

Hozzáférés

-0.45

Personensuche

-0.45

 MongoClient

-0.45

POSITIVE LOGITS

 humaine

0.42

 CURIAM

0.42

enjoy

0.41

 techniczne

0.40

 enjoyment

0.40

 attractive

0.40

attractive

0.39

baliknya

0.39

 beautiful

0.39

 pięk

0.39

Activations Density 0.012%

positive adjectives and phrases related to performance and experiences

positive descriptions and enjoyment

No Comments

No Known Activations