INDEX

Explanations

descriptions related to negative attitudes or skepticism

oai_token-act-pair · gpt-3.5-turbo

discussions of cynicism and cynical attitudes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

acid

-0.80

ĸļ

-0.78

erto

-0.75

hner

-0.72

upon

-0.71

ymes

-0.69

edited

-0.69

Hall

-0.68

avery

-0.66

ŃĶ

-0.65

POSITIVE LOGITS

cyn

1.01

 observer

0.81

 cynicism

0.79

 ploy

0.76

Cyn

0.74

 overlook

0.74

 cynical

0.72

 bastard

0.71

psy

0.70

 twist

0.69

Activations Density 0.014%

descriptions related to negative attitudes or skepticism

discussions of cynicism and cynical attitudes

No Comments

No Known Activations