INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

lift

-0.76

tight

-0.69

ilee

-0.68

relation

-0.67

hig

-0.64

neys

-0.64

riel

-0.64

":"/

-0.63

vati

-0.63

cases

-0.61

POSITIVE LOGITS

avorite

0.73

ð

0.71

¥ŀ

0.65

gypt

0.63

 hell

0.62

Bav

0.62

oise

0.62

Zh

0.62

 Sina

0.61

 besie

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.