INDEX

Explanations

references to the brand "Lego"

oai_token-act-pair · gpt-3.5-turbo

references to the Lego brand and its products

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

yip

-0.85

ysis

-0.79

bern

-0.73

pressed

-0.72

sweet

-0.71

uncture

-0.70

ebin

-0.69

ntil

-0.69

terness

-0.68

robe

-0.67

POSITIVE LOGITS

 Dimensions

0.92

 Ideas

0.92

 LEGO

0.89

 bricks

0.87

 Lego

0.87

Â®

0.76

 Theft

0.74

 Minecraft

0.73

 Batman

0.72

ota

0.70

Activations Density 0.023%

references to the brand "Lego"

references to the Lego brand and its products

No Comments

No Known Activations