INDEX

Explanations

text related to technical and medical imaging terms

oai_token-act-pair · gpt-3.5-turbo

specific visual elements or illustrations in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 addict

-0.68

answered

-0.67

gress

-0.66

abol

-0.64

dule

-0.63

 listeners

-0.61

nesia

-0.61

ophobic

-0.60

netflix

-0.60

antic

-0.60

POSITIVE LOGITS

Courtesy

1.09

 Courtesy

1.08

tesy

1.03

 photographed

1.00

 Photo

0.96

 Photographer

0.95

 PHOTO

0.93

 courtesy

0.93

 Illust

0.93

 Â©

0.92

Activations Density 0.512%

text related to technical and medical imaging terms

specific visual elements or illustrations in the text

No Comments

No Known Activations