INDEX

Explanations

instances of the word "attitude" with varying degrees of intensity

oai_token-act-pair · gpt-3.5-turbo

mentions of specific attitudes or perspectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

enegger

-0.89

icles

-0.87

icle

-0.75

dry

-0.71

idden

-0.67

oval

-0.67

baum

-0.66

anz

-0.66

schild

-0.65

Interstitial

-0.65

POSITIVE LOGITS

 toward

1.32

 towards

1.28

 attitude

1.12

 Towards

1.03

 attitudes

0.88

Tow

0.82

ysis

0.78

 stance

0.76

 hostility

0.73

uation

0.73

Activations Density 0.038%

instances of the word "attitude" with varying degrees of intensity

mentions of specific attitudes or perspectives

No Comments

No Known Activations