INDEX

Explanations

conjunctions like 'and' or 'or' connecting phrases with contrasting effects

oai_token-act-pair · gpt-3.5-turbo

comparative phrases discussing the interplay of positive and negative aspects

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

£ı

-0.95

ĸļ

-0.95

eday

-0.81

ItemThumbnailImage

-0.79

ESE

-0.75

Ħ¢

-0.73

monary

-0.72

-0.71

dayName

-0.71

ģĸ

-0.70

POSITIVE LOGITS

bad

1.19

 evil

1.05

bad

1.05

 worst

1.04

 drawbacks

1.02

 downside

1.01

 disapproval

1.01

 disappoint

1.00

 unpleasant

0.99

 harmful

0.98

Activations Density 0.271%

conjunctions like 'and' or 'or' connecting phrases with contrasting effects

comparative phrases discussing the interplay of positive and negative aspects

No Comments

No Known Activations