INDEX

Explanations

interactions and emotional responses between characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

Jîn

-0.70

 AssemblyCulture

-0.68

 Италијани

-0.65

majánló

-0.63

ロウィン

-0.63

rungsseite

-0.63

chieht

-0.61

oredCriteria

-0.61

 mío

-0.59

BibitemShut

-0.59

POSITIVE LOGITS

 unaware

0.40

Aware

0.37

 onge

0.33

aware

0.30

 Aware

0.30

 Unbe

0.30

 unbe

0.29

 aware

0.28

为

0.28

Activations Density 0.192%

interactions and emotional responses between characters

No Comments

No Known Activations