INDEX

Explanations

instances where there is a willingness to do something

oai_token-act-pair · gpt-3.5-turbo

expressions of willingness or unwillingness to take action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

icle

-0.87

icles

-0.76

inas

-0.74

agraph

-0.70

adish

-0.70

 Regions

-0.69

 Sections

-0.67

 Panther

-0.66

 Anthem

-0.65

NCT

-0.64

POSITIVE LOGITS

 willingness

1.05

 unwillingness

0.93

yip

0.91

 guiActiveUn

0.90

 attitude

0.88

ï¸

0.86

terday

0.78

 willingly

0.76

 stance

0.75

 reluctance

0.74

Activations Density 0.006%

instances where there is a willingness to do something

expressions of willingness or unwillingness to take action

No Comments

No Known Activations