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

racuse

-0.88

creen

-0.88

zed

-0.78

chrom

-0.77

ridges

-0.76

puters

-0.75

cess

-0.75

hattan

-0.72

":""},{"

-0.69

zers

-0.68

POSITIVE LOGITS

ãĥ¡

0.78

EGIN

0.72

BIP

0.69

ONSORED

0.69

ORN

0.69

CDC

0.68

ALSE

0.67

 Pastebin

0.66

TAM

0.66

AFP

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.