INDEX

Explanations

references to the concept of honor

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

 Vorstellung

-0.57

 Decke

-0.48

 Menschheit

-0.46

 Entstehung

-0.46

 Notwendigkeit

-0.45

 Substanz

-0.44

 powsta

-0.43

 pomysł

-0.42

 Teilnahme

-0.42

 fantasmas

-0.41

POSITIVE LOGITS

 honor

0.91

honor

0.88

 Honor

0.86

Honor

0.85

 Harbour

0.80

 Anim

0.79

 honour

0.79

 HONOR

0.77

:✨

0.77

 للمعارف

0.76

Activations Density 0.249%

references to the concept of honor

No Comments

No Known Activations