INDEX

Explanations

adjectives that emphasize magnitude or degree

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

following "more" or "very"

np_acts-logits-general · gemini-2.0-flash

intensifiers before adjectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

 avoient

-0.36

MigrationBuilder

-0.34

MemoryWarning

-0.33

 Wikiseite

-0.33

 personnalisée

-0.32

 plazo

-0.32

 personnalisées

-0.32

 maksi

-0.32

 résult

-0.32

 sév

-0.31

POSITIVE LOGITS

 unſ

0.58

✭✭

0.57

 adaptable

0.57

twimg

0.56

 kasarigan

0.56

 aware

0.56

AddHtmlAttribute

0.54

 communicative

0.53

 blest

0.52

 MyBatis

0.51

Activations Density 0.240%

adjectives that emphasize magnitude or degree

following "more" or "very"

intensifiers before adjectives

No Comments

No Known Activations