INDEX

Explanations

words related to medical and health conditions

oai_token-act-pair · gpt-3.5-turbo

words associated with specific individuals or identities

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

iating

-0.71

iates

-0.70

lied

-0.70

 Tablet

-0.69

 "$:/

-0.67

 Defender

-0.65

des

-0.64

ista

-0.63

itatively

-0.63

izational

-0.62

POSITIVE LOGITS

isy

1.14

\\\\\\\\\\\\\\\\

0.91

vous

0.80

abeth

0.75

querade

0.71

 Cage

0.71

Ô

0.70

achus

0.67

bda

0.66

cean

0.64

Activations Density 0.025%

words related to medical and health conditions

words associated with specific individuals or identities

No Comments

No Known Activations