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

DragonMagazine

-1.01

tackle

-0.76

uese

-0.74

twitch

-0.68

ftime

-0.67

rame

-0.67

isSpecialOrderable

-0.67

worthiness

-0.66

ĸļ

-0.65

"""

-0.64

POSITIVE LOGITS

 Tours

0.75

 Solutions

0.68

 Nightmares

0.67

acles

0.65

cat

0.64

 Procedures

0.63

Ong

0.63

olyn

0.62

RIS

0.62

 Sear

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.