INDEX

Explanations

references to things being simple, clear, or evident

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "plain."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

glomer

-0.78

otos

-0.76

hyde

-0.74

vez

-0.72

uana

-0.71

yip

-0.70

ofi

-0.69

allery

-0.68

waukee

-0.68

alez

-0.67

POSITIVE LOGITS

plain

1.32

 plain

1.15

text

1.06

ãĤ¦ãĤ¹

0.80

\\\\\\\\

0.76

ified

0.75

cloth

0.75

 Plain

0.74

wallet

0.73

 vanilla

0.71

Activations Density 0.008%

references to things being simple, clear, or evident

references to the concept of "plain."

No Comments

No Known Activations