INDEX

Explanations

captions for images

oai_token-act-pair · gpt-3.5-turbo

references to images and credits in articles

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

ãĥĺ

-0.78

 secretaries

-0.67

adobe

-0.67

 elev

-0.65

gam

-0.64

anmar

-0.62

 Mahjong

-0.62

 polymorph

-0.62

puter

-0.61

 resear

-0.61

POSITIVE LOGITS

CBC

0.84

 Photo

0.81

ONSORED

0.79

AFP

0.77

 IMAGES

0.77

Photo

0.77

ANI

0.77

 Courtesy

0.77

 Everett

0.74

 Grip

0.73

Activations Density 0.203%

captions for images

references to images and credits in articles

No Comments

No Known Activations