INDEX

Explanations

mentions of and references to cameras

oai_token-act-pair · gpt-3.5-turbo

references to a camera

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ensable

-0.88

emonic

-0.87

doms

-0.75

luent

-0.75

heit

-0.74

inence

-0.71

wegian

-0.70

sted

-0.69

enberg

-0.69

itable

-0.68

POSITIVE LOGITS

 Camer

1.23

 camera

1.14

 cameras

1.10

 Camera

1.03

Camera

0.96

 shutter

0.94

 pans

0.93

 Nikon

0.92

 camer

0.92

 tripod

0.90

Activations Density 0.016%

mentions of and references to cameras

references to a camera

No Comments

No Known Activations