INDEX

Explanations

references to dining experiences and food offerings

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

food, service, value, drama, features

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

Diweddarwch

-0.39

 adecuada

-0.35

 Meksiku

-0.35

……”

-0.34

 alimentation

-0.33

 yeterli

-0.32

 aplicable

-0.32

 instrumento

-0.32

..........

-0.30

 Municipio

-0.30

POSITIVE LOGITS

ſcher

0.54

 ſein

0.54

mbggenerated

0.53

⤹

0.53

 stunning

0.53

__*/

0.53

ftance

0.52

 comfy

0.52

 laſſen

0.51

AndEndTag

0.51

Activations Density 0.006%

references to dining experiences and food offerings

food, service, value, drama, features

No Comments

No Known Activations