INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĵĺ

-0.80

EStreamFrame

-0.78

bern

-0.76

 Delay

-0.74

ĨĴ

-0.71

isites

-0.69

imester

-0.66

aucus

-0.66

 Chosen

-0.65

abet

-0.64

POSITIVE LOGITS

agos

0.76

HMS

0.70

 Nikon

0.69

0.64

RAF

0.64

 spraying

0.61

waters

0.61

 Kitty

0.61

tes

0.61

iverpool

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.