INDEX

Explanations

occurrences of the word "listed" and its variations, indicating a focus on references to items or subjects that are categorized or mentioned in lists

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

inclusion in lists

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

cvtColor

-0.58

 Cromwell

-0.57

 Cronin

-0.56

cmath

-0.56

 Garnett

-0.54

April

-0.53

 McFarland

-0.52

newPassword

-0.52

umenical

-0.51

 polymorph

-0.50

POSITIVE LOGITS

 listed

1.08

listed

0.92

 Listed

0.91

Listed

0.91

 listing

0.86

 liste

0.79

 list

0.78

 listings

0.77

 Listing

0.77

 LIST

0.76

Activations Density 0.022%

occurrences of the word "listed" and its variations, indicating a focus on references to items or subjects that are categorized or mentioned in lists

inclusion in lists

No Comments

No Known Activations