INDEX

Explanations

names of famous historical figures, especially those related to World War II and mathematics

oai_token-act-pair · gpt-3.5-turbo

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

cript

-0.82

ilver

-0.77

peak

-0.76

apeshifter

-0.75

 sclerosis

-0.72

akening

-0.69

pell

-0.68

BOOK

-0.67

creen

-0.67

raq

-0.66

POSITIVE LOGITS

sson

1.07

 Hitler

1.00

0.98

 Schwarzenegger

0.96

 Wilhelm

0.94

erick

0.85

uez

0.85

 Hitchcock

0.84

 Franz

0.83

 Canaver

0.83

Activations Density 0.055%

names of famous historical figures, especially those related to World War II and mathematics

No Comments

No Known Activations