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

achus

-0.75

itability

-0.66

 motors

-0.65

ashtra

-0.65

uese

-0.64

asuring

-0.64

hammer

-0.63

arnaev

-0.63

Downloadha

-0.62

itol

-0.61

POSITIVE LOGITS

 schizophren

0.80

 Whitman

0.72

ctor

0.72

 Beckham

0.71

 Dalai

0.70

esc

0.70

 Replay

0.69

 Crawford

0.67

 Hoover

0.64

 Complex

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.