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

atform

-0.86

artz

-0.68

 inherit

-0.63

urally

-0.63

 reconnect

-0.63

 breadth

-0.62

ounty

-0.62

eways

-0.61

iazep

-0.61

ecast

-0.60

POSITIVE LOGITS

 Lucia

0.90

usha

0.72

Bride

0.69

 Morg

0.69

Ã¨re

0.68

Ã©e

0.68

Ã¨

0.67

 Princ

0.64

cel

0.64

Cec

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.