INDEX

Explanations

words related to the Wired Houlton entity, including descriptions and offerings

oai_token-act-pair · gpt-3.5-turbo

instances of the term "Wired."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ouri

-0.97

naire

-0.86

gary

-0.76

buster

-0.76

othal

-0.75

itol

-0.74

oresc

-0.73

igating

-0.72

azon

-0.71

ptive

-0.70

POSITIVE LOGITS

nesday

0.82

 EDITION

0.80

 Printed

0.74

æ©Ł

0.73

sung

0.68

yssey

0.67

 Haram

0.66

apons

0.65

DIT

0.65

ãĤ¦ãĤ¹

0.65

Activations Density 0.081%

words related to the Wired Houlton entity, including descriptions and offerings

instances of the term "Wired."

No Comments

No Known Activations