INDEX

Explanations

mentions of sin

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "sin."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

æ©Ł

-0.76

assetsadobe

-0.69

ership

-0.69

 Pione

-0.66

 downed

-0.66

IFE

-0.65

dropping

-0.64

ATER

-0.62

PsyNetMessage

-0.62

DragonMagazine

-0.61

POSITIVE LOGITS

sin

1.12

ned

0.99

Sin

0.92

estro

0.92

atra

0.90

uous

0.88

ning

0.87

 Sins

0.87

isphere

0.86

 sinful

0.82

Activations Density 0.012%

mentions of sin

occurrences of the word "sin."

No Comments

No Known Activations