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

 subpoena

-0.78

heses

-0.69

olan

-0.69

 waivers

-0.66

 Bhar

-0.63

grading

-0.63

 Dame

-0.62

 Leopard

-0.61

apixel

-0.61

 Allaah

-0.61

POSITIVE LOGITS

ilib

0.71

low

0.66

strap

0.65

teenth

0.65

 mould

0.64

thing

0.64

çī

0.62

LOW

0.61

 shallow

0.60

UME

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.