INDEX

Explanations

adjectives that convey positivity and significance

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

positive adjectives followed by nouns

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

ioutil

-0.57

 hâte

-0.48

 those

-0.48

LookAnd

-0.47

 eivät

-0.45

 aveug

-0.45

 SimpleName

-0.42

 Preise

-0.42

 générale

-0.41

 payloads

-0.41

POSITIVE LOGITS

 acestei

0.55

 wonderful

0.49

DebuggerNonUser

0.49

 acestui

0.49

 Этот

0.47

 Denna

0.46

 Dieses

0.45

 exploratory

0.45

 Эта

0.44

 exciting

0.43

Activations Density 0.027%

adjectives that convey positivity and significance

positive adjectives followed by nouns

No Comments

No Known Activations