INDEX

Explanations

references to clubs or groups related to social activities

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

"club"

np_acts-logits-general · gemini-2.0-flash

sports club, community club

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Respuesta

-0.57

 Respuesta

-0.56

 mensagens

-0.51



-0.51

 gingerbread

-0.49

 Answers

-0.48

 Message

-0.48

 aéri

-0.48

 paziente

-0.48

 mariée

-0.48

POSITIVE LOGITS

Club

0.95

 CLUB

0.92

 club

0.91

 Club

0.89

 Clubs

0.86

 clubs

0.84

Clubs

0.84

CLUB

0.82

club

0.81

clubs

0.79

Activations Density 0.102%

references to clubs or groups related to social activities

"club"

sports club, community club

No Comments

No Known Activations