INDEX

Explanations

mentions of Formula 1 racing drivers and events

oai_token-act-pair · gpt-3.5-turbo

references to notable figures and their performances in a competitive context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ormons

-0.86

ibrary

-0.86

adelphia

-0.86

adish

-0.85

Zup

-0.81

 Mormons

-0.79

 Hallow

-0.79

gat

-0.78

yden

-0.76

 Garrison

-0.74

POSITIVE LOGITS

 McLaren

1.89

 Ferrari

1.83

 Renault

1.70

FIA

1.68

 Mercedes

1.68

 Motorsport

1.66

 tyres

1.65

 tyre

1.62

 Formula

1.57

 racing

1.50

Activations Density 0.534%

mentions of Formula 1 racing drivers and events

references to notable figures and their performances in a competitive context

No Comments

No Known Activations