INDEX

Explanations

references to relationships and friendships

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

preceding "friends" or "friend"

np_acts-logits-general · gemini-2.0-flash

closest friends

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

>[]

-0.49

JsonValue

-0.45

躇

-0.44

 desta

-0.44

kac

-0.42

isele

-0.42

tructuring

-0.42

 cdti

-0.41

 sement

-0.41

 couvert

-0.40

POSITIVE LOGITS

 friends

2.95

 friend

2.77

friend

2.45

friends

2.44

 Friends

2.28

Friends

2.27

Friend

2.26

 FRIENDS

2.24

 Friend

2.23

 FRIEND

2.17

Activations Density 0.248%

references to relationships and friendships

preceding "friends" or "friend"

closest friends

No Comments

No Known Activations