INDEX

Explanations

mentions of the word "Brit" followed by a number or a variation like "Britons"

oai_token-act-pair · gpt-3.5-turbo

references to British individuals or the term "Brit."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

REDACTED

-0.79

xual

-0.79

cember

-0.78

externalActionCode

-0.74

LEASE

-0.72

mble

-0.70

CVE

-0.70

 recess

-0.70

CONCLUS

-0.68

TPPStreamerBot

-0.67

POSITIVE LOGITS

ney

1.06

ishers

1.03

 Brit

0.95

 Hume

0.93

isher

0.87

neys

0.85

oise

0.82

 Spears

0.78

odi

0.78

 Airways

0.78

Activations Density 0.009%

mentions of the word "Brit" followed by a number or a variation like "Britons"

references to British individuals or the term "Brit."

No Comments

No Known Activations