INDEX

Explanations

phrases related to being supportive or supportiveness

oai_token-act-pair · gpt-3.5-turbo

references to supportive behaviors or attitudes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

hunt

-0.79

resp

-0.75

ëĭ

-0.70

hun

-0.70

gob

-0.69

hig

-0.69

uld

-0.69

tein

-0.68

jong

-0.67

oqu

-0.67

POSITIVE LOGITS

 supportive

1.23

 Supports

1.03

hetical

0.83

 supports

0.81

 endorsements

0.80

heses

0.79

 support

0.78

 SUPPORT

0.77

 backers

0.76

 encour

0.76

Activations Density 0.008%

phrases related to being supportive or supportiveness

references to supportive behaviors or attitudes

No Comments

No Known Activations