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

 Observatory

-0.71

pson

-0.69

 Travels

-0.64

 messenger

-0.64

isine

-0.63

wright

-0.63

ridges

-0.62

bridge

-0.61

OY

-0.61

yip

-0.61

POSITIVE LOGITS

 there

0.92

ļéĨĴ

0.86

there

0.76

 THERE

0.67

vard

0.63

rompt

0.62

:\

0.61

ŃĶ

0.60

abl

0.60

BSD

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.