INDEX

Explanations

references to historical and literary works, particularly related to translations or digitizations

oai_token-act-pair · gpt-3.5-turbo

references to historical or scholarly works and authentic translations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ndra

-0.91

twitch

-0.91

deals

-0.82

amacare

-0.79

>>\

-0.78

à¼

-0.75

ggies

-0.75

aturday

-0.74

paces

-0.74

ggle

-0.73

POSITIVE LOGITS

 manuscript

1.91

 manuscripts

1.67

 reprint

1.49

 editions

1.41

 edition

1.38

 translation

1.37

uscript

1.35

 unpublished

1.32

 paperback

1.32

 pamphlet

1.31

Activations Density 0.314%

references to historical and literary works, particularly related to translations or digitizations

references to historical or scholarly works and authentic translations

No Comments

No Known Activations