INDEX

Explanations

locations or directions

oai_token-act-pair · gpt-3.5-turbo

references to visual or spatial descriptions

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

heit

-0.76

 depended

-0.74

 maintains

-0.70

 insured

-0.68

 maintained

-0.67

 insisted

-0.66

 ensured

-0.65

 exclus

-0.63

untary

-0.63

 guaranteed

-0.63

POSITIVE LOGITS

ideos

0.84

Picture

0.83

photos

0.78

 screenshots

0.78

 pictures

0.76

 afar

0.75

 pics

0.74

 footprints

0.72

 VIDEO

0.72

 CCTV

0.71

Activations Density 0.479%

locations or directions

references to visual or spatial descriptions

No Comments

No Known Activations