INDEX

Explanations

terms related to lighting

oai_token-act-pair · gpt-3.5-turbo

references to lighting conditions

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

gery

-0.77

essor

-0.77

clair

-0.77

berus

-0.77

sburgh

-0.76

irlf

-0.76

llah

-0.72

ocide

-0.72

eday

-0.71

apego

-0.70

POSITIVE LOGITS

 torches

1.02

 bulbs

1.00

 bulb

0.97

 fixtures

0.96

 lighting

0.95

 candles

0.89

 lamps

0.87

lights

0.84

 fixture

0.83

 lights

0.80

Activations Density 0.013%

terms related to lighting

references to lighting conditions

No Comments

No Known Activations