INDEX

Explanations

mentions and variations of the word "curse."

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

curse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

APOLIS

-0.46

 للمعارف

-0.44

GOTREF

-0.41

assertAll

-0.40

Inet

-0.37

PINES

-0.36

 typelib

-0.35

WAM

-0.35

RIO

-0.34

eabouts

-0.34

POSITIVE LOGITS

curse

0.76

Curse

0.69

 Curse

0.67

 curse

0.66

cursed

0.60

TypedDataSet

0.59

Cursed

0.58

 curses

0.56

 cursing

0.56

 Cursed

0.55

Activations Density 0.007%

mentions and variations of the word "curse."

curse

No Comments

No Known Activations