INDEX

Explanations

words related to controversy or sensitive topics

oai_token-act-pair · gpt-3.5-turbo

phrases related to donations or support

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 apprentices

-0.74

 endeavour

-0.73

 volunt

-0.72

lling

-0.71

ilibrium

-0.71

avering

-0.70

uly

-0.67

 principally

-0.67

 endeav

-0.66

 consecut

-0.66

POSITIVE LOGITS

Anyway

1.36

Advertisement

1.19

Seriously

1.10

UPDATE

1.00

 Anyway

0.96

Liter

0.95

Yep

0.94

HAHAHAHA

0.93

Bonus

0.92

PHOTOS

0.92

Activations Density 0.678%

words related to controversy or sensitive topics

phrases related to donations or support

No Comments

No Known Activations