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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tanz

-0.67

Ran

-0.66

phys

-0.62

 Synd

-0.62

Narr

-0.62

Barn

-0.62

 Meyer

-0.62

Sir

-0.61

Events

-0.61

 Tate

-0.61

POSITIVE LOGITS

andals

0.83

oshenko

0.79

OPLE

0.72

hovah

0.71

duty

0.71

uctor

0.70

elf

0.69

dinand

0.68

yip

0.68

ometimes

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.