INDEX

Explanations

direct speech quotations in written text

oai_token-act-pair · gpt-3.5-turbo

direct quotes from individuals expressing opinions or statements

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stakes

-0.67

ãĤ¼

-0.62

Contents

-0.60

brance

-0.60

 PHOTO

-0.60

 coincides

-0.59

icycle

-0.58

isin

-0.57

veh

-0.57

agram

-0.57

POSITIVE LOGITS

 emphatically

0.73

 bluntly

0.72

lip

0.71

etheless

0.66

 affirmative

0.65

 sarcast

0.64

"[

0.64

yes

0.64

:"

0.64

 hesitation

0.63

Activations Density 0.168%

direct speech quotations in written text

direct quotes from individuals expressing opinions or statements

No Comments

No Known Activations