INDEX

Explanations

phrases indicating a negation or contradiction

oai_token-act-pair · gpt-3.5-turbo

negations or the word "not"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

velt

-0.73

æ©

-0.71

Ĥİ

-0.68

ushes

-0.67

å¥

-0.67

 Tours

-0.67

 Laws

-0.66

ãĤ»

-0.65

PsyNetMessage

-0.65

 Tens

-0.65

POSITIVE LOGITS

 necessarily

1.23

icably

1.23

icable

1.10

yet

1.06

hin

1.05

eworthy

1.05

epad

1.04

 permitted

1.00

 allowed

0.99

orious

0.95

Activations Density 0.141%

phrases indicating a negation or contradiction

negations or the word "not"

No Comments

No Known Activations