INDEX

Explanations

phrases addressing or referring to the reader

oai_token-act-pair · gpt-3.5-turbo

references to the audience or group of people being addressed

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

orp

-0.67

 Heist

-0.62

 Samoa

-0.60

ayne

-0.60

 Faul

-0.60

 Hick

-0.59

 Racial

-0.59

entimes

-0.59

 Canaver

-0.58

 Newport

-0.58

POSITIVE LOGITS

tub

1.17

 guys

1.05

 gentlemen

0.83

RS

0.83

 yourselves

0.83

azeera

0.76

 know

0.74

TH

0.73

TER

0.72

é»Ĵ

0.71

Activations Density 0.037%

phrases addressing or referring to the reader

references to the audience or group of people being addressed

No Comments

No Known Activations