INDEX

Explanations

the word "not" followed by a description or comparison

oai_token-act-pair · gpt-3.5-turbo

negations and contrasting statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

tein

-0.80

FAQ

-0.73

 Crate

-0.70

hower

-0.67

ARE

-0.67

velt

-0.66

irt

-0.65

ESE

-0.63

ye

-0.62

undrum

-0.62

POSITIVE LOGITS

 necessarily

1.47

epad

1.19

icable

1.06

 bothering

1.02

icably

0.99

 bothered

0.91

eworthy

0.90

ifies

0.89

 bother

0.88

 merely

0.88

Activations Density 0.066%

the word "not" followed by a description or comparison

negations and contrasting statements

No Comments

No Known Activations