INDEX

Explanations

phrases related to opinions or attitudes of individuals

oai_token-act-pair · gpt-3.5-turbo

expressions of emotional responses and attitudes towards situations or events

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 miracle

-0.70

utters

-0.65

 Loaded

-0.63

 defic

-0.59

 Basics

-0.58

Database

-0.58

 dummy

-0.58

iller

-0.55

 absurdity

-0.54

ingham

-0.54

POSITIVE LOGITS

 toward

1.42

 about

1.41

about

1.35

 ABOUT

1.35

 towards

1.33

 About

1.18

About

1.10

 regarding

1.09

 Regarding

0.96

 concerning

0.92

Activations Density 0.321%

phrases related to opinions or attitudes of individuals

expressions of emotional responses and attitudes towards situations or events

No Known Activations