INDEX

Explanations

references to cameras or camera-related actions/objects

oai_token-act-pair · gpt-3.5-turbo

references to cameras

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

emonic

-0.83

hee

-0.77

hend

-0.74

ensical

-0.72

heit

-0.72

querque

-0.71

edin

-0.69

tions

-0.67

wegian

-0.66

dict

-0.66

POSITIVE LOGITS

 Camer

1.00

 cameras

0.92

 shutter

0.91

Lens

0.88

 microphones

0.88

 Nikon

0.86

 lens

0.86

 camera

0.86

 pans

0.85

 Operator

0.85

Activations Density 0.022%

references to cameras or camera-related actions/objects

references to cameras

No Comments

No Known Activations