INDEX

Explanations

frequent words and pronouns indicating personal connection or ownership

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

 émission

-0.76

 article

-0.75

 answer

-0.75

 argument

-0.74

 artist

-0.73

 idea

-0.72

 album

-0.71

 alternative

-0.71

 array

-0.71

 odeur

-0.71

POSITIVE LOGITS

extrémité

0.68

épaisseur

0.60

équipement

0.60

épreuve

0.60

extérieur

0.59

essentiel

0.59

étoit

0.58

indépendance

0.57

égard

0.57

échange

0.57

Activations Density 0.088%

frequent words and pronouns indicating personal connection or ownership

No Comments

No Known Activations