INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

istas

-0.78

Spot

-0.68

ters

-0.67

batch

-0.65

ista

-0.63

tering

-0.62

worn

-0.62

 hairc

-0.62

roots

-0.61

urgical

-0.61

POSITIVE LOGITS

agra

0.74

urai

0.72

aga

0.68

asus

0.68

avorite

0.67

 Gork

0.66

 Dart

0.64

TCU

0.62

Yug

0.62

REE

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.