INDEX

Explanations

references to the North Korean leader Kim Jong Un

oai_token-act-pair · gpt-3.5-turbo

references to North Korean leader Kim Jong Un

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ONY

-0.78

VID

-0.75

 GOODMAN

-0.74

rawdownloadcloneembedreportprint

-0.72

JECT

-0.71

WN

-0.70

EQ

-0.70

ãĤ©

-0.69

 discrep

-0.67

 Racer

-0.66

POSITIVE LOGITS

 Jinping

0.98

 Lumpur

0.82

Jr

0.76

 ascended

0.76

bart

0.75

 Himself

0.75

 himself

0.73

Jr

0.72

 confid

0.72

wei

0.72

Activations Density 0.075%

references to the North Korean leader Kim Jong Un

references to North Korean leader Kim Jong Un

No Comments

No Known Activations