INDEX

Explanations

phrases or words related to things not being tightly secured or controlled

oai_token-act-pair · gpt-3.5-turbo

the word "loose" in various contexts

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

olan

-0.72

Īè

-0.71

VERTISEMENT

-0.67

JUST

-0.67

 Prize

-0.66

 Appeals

-0.64

Pax

-0.64

ãĤ´ãĥ³

-0.64

ivals

-0.64

Reviewed

-0.63

POSITIVE LOGITS

 loose

1.07

leaf

0.97

weed

0.85

cloth

0.81

packed

0.80

luster

0.80

bound

0.75

tail

0.73

tails

0.73

lege

0.73

Activations Density 0.008%

phrases or words related to things not being tightly secured or controlled

the word "loose" in various contexts

No Comments

No Known Activations