INDEX

Explanations

the occurrence of the verb "used" and its variations in the text

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

describing use or purpose

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

TokenNameDOT

-0.51

stonia

-0.47

obe

-0.45

ilions

-0.41

uarts

-0.41

nthetic

-0.41

ayet

-0.41

Kanpo

-0.41

quo

-0.40

obey

-0.40

POSITIVE LOGITS

 Used

0.68

 used

0.65

Used

0.61

 verwendeten

0.57

 USED

0.53

 CreateTagHelper

0.52

 utilisés

0.51

 použit

0.50

 brukt

0.50

cydow

0.49

Activations Density 0.077%

the occurrence of the verb "used" and its variations in the text

describing use or purpose

No Comments

No Known Activations