INDEX

Explanations

adverbs that indicate frequency or likelihood

oai_token-act-pair · gpt-3.5-turbo

adverbs and modifiers indicating frequency or likelihood

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 adamant

-0.66

 careful

-0.63

ASE

-0.62

 gazing

-0.58

 Gothic

-0.58

 tweeting

-0.58

 Azerbai

-0.58

 Tray

-0.58

 arguing

-0.57

ARCH

-0.57

POSITIVE LOGITS

 belongs

1.31

 reflects

1.30

 exists

1.28

 occurs

1.27

 revolves

1.26

 corresponds

1.24

 coincides

1.24

 represents

1.23

 happens

1.23

 involves

1.22

Activations Density 0.279%

adverbs that indicate frequency or likelihood

adverbs and modifiers indicating frequency or likelihood

No Comments

No Known Activations