INDEX

Explanations

mentions of specific names, particularly "Elon Musk" and "Theodore"

oai_token-act-pair · gpt-3.5-turbo

mentions of influential figures, particularly Elon Musk and Theodore Roosevelt

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

maid

-0.79

lez

-0.72

eding

-0.71

Torrent

-0.71

scenes

-0.69

esome

-0.69

united

-0.67

etz

-0.67

hops

-0.67

alez

-0.66

POSITIVE LOGITS

 Musk

1.32

 Elon

0.87

itability

0.81

uates

0.81

 Roosevelt

0.80

 Brock

0.77

 Watts

0.76

ufact

0.76

ancock

0.74

 Manning

0.73

Activations Density 0.025%

mentions of specific names, particularly "Elon Musk" and "Theodore"

mentions of influential figures, particularly Elon Musk and Theodore Roosevelt

No Comments

No Known Activations