INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Afric

-0.76

HMS

-0.68

 CENT

-0.64

trak

-0.63

umerable

-0.62

 Somers

-0.62

ioxide

-0.61

ighed

-0.61

Course

-0.61

ļéĨĴ

-0.60

POSITIVE LOGITS

 drawer

0.70

ni

0.66

tub

0.66

stakes

0.65

 prest

0.64

 contrace

0.63

 nodd

0.63

 trending

0.63

 bribe

0.62

 fung

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.