INDEX

Explanations

negative qualities and criticisms related to arrogance and deceitfulness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 الرياضيه

-0.68

elemField

-0.57

Rüyada

-0.54

 Kebijakan

-0.52

OGND

-0.52

 iſt

-0.51

 ujednoznacz

-0.51

sena

-0.51

 čierna

-0.51

IntoConstraints

-0.50

POSITIVE LOGITS

 jerk

0.92

 jerks

0.85

 douche

0.83

 prick

0.77

 idiot

0.75

 moron

0.74

 imbec

0.73

 asshole

0.73

 scum

0.69

 loser

0.69

Activations Density 0.508%

negative qualities and criticisms related to arrogance and deceitfulness

No Comments

No Known Activations