INDEX

Explanations

references and mentions of the brand LEGO and related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of LEGO products and related terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ktop

-0.76

rine

-0.74

robe

-0.72

sweet

-0.71

pher

-0.70

terness

-0.70

yip

-0.69

uncture

-0.68

selection

-0.67

ysis

-0.67

POSITIVE LOGITS

 Dimensions

1.15

 Ideas

1.06

 bricks

1.00

 Lego

0.94

 LEGO

0.91

 Turtles

0.86

 Batman

0.85

Â®

0.84

 Mansion

0.83

 creations

0.80

Activations Density 0.066%

references and mentions of the brand LEGO and related terms

mentions of LEGO products and related terms

No Comments

No Known Activations