INDEX

Explanations

phrases related to sporting venues or stadiums

oai_token-act-pair · gpt-3.5-turbo

references to stadiums

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

thodox

-0.82

dit

-0.77

yan

-0.74

leans

-0.70

bart

-0.69

 Leilan

-0.67

bery

-0.65

istries

-0.64

uscript

-0.63

Dist

-0.63

POSITIVE LOGITS

 stadiums

1.12

 stadium

1.06

adium

0.81

 Stadium

0.75

 auditor

0.71

 seating

0.71

 amph

0.71

 announcer

0.71

 entrances

0.69

 arena

0.69

Activations Density 0.024%

phrases related to sporting venues or stadiums

references to stadiums

No Comments

No Known Activations