INDEX

Explanations

mentions of or related to autographs, autographed items, and autograph signings

oai_token-act-pair · gpt-3.5-turbo

references to signatures and autographs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

lihood

-0.88

forth

-0.74

 Forbidden

-0.68

imaru

-0.66

FORE

-0.66

BLE

-0.66

ITED

-0.66

 Penet

-0.63

 Soda

-0.63

nesday

-0.61

POSITIVE LOGITS

ographs

1.59

ograph

1.57

ographed

1.36

opsy

1.33

umn

1.32

ocom

1.27

ograp

1.22

ocrat

1.22

onomous

1.20

omatic

1.15

Activations Density 0.018%

mentions of or related to autographs, autographed items, and autograph signings

references to signatures and autographs

No Comments

No Known Activations