INDEX

Explanations

specific phrases or sentences on physical signs or stickers

oai_token-act-pair · gpt-3.5-turbo

references to signs, banners, or labels that contain written messages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

unker

-0.75

kus

-0.71

olit

-0.70

erker

-0.68

emort

-0.65

ecast

-0.65

anu

-0.65

ester

-0.64

orget

-0.64

odcast

-0.64

POSITIVE LOGITS

"#

1.18

"+

0.95

"$

0.91

"@

0.85

"%

0.82

"_

0.78

 logos

0.76

:"

0.75

""

0.74

"'

0.74

Activations Density 0.274%

specific phrases or sentences on physical signs or stickers

references to signs, banners, or labels that contain written messages

No Comments

No Known Activations