INDEX

Explanations

variations of the word "Hon" and associated honorary titles or references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

twimg

-0.46

 matters

-0.40

 nuages

-0.38

?>">

-0.36

 дописавши

-0.35

 mosca

-0.35

matters

-0.34

afone

-0.34

ñado

-0.34

matter

-0.34

POSITIVE LOGITS

hon

1.19

Hon

1.06

hon

1.05

Hon

1.04

HON

0.89

HON

0.86

 HONORABLE

0.71

 Honourable

0.71

 honourable

0.71

 Honorable

0.70

Activations Density 0.232%

variations of the word "Hon" and associated honorary titles or references

No Comments

No Known Activations