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

estamp

-0.83

ional

-0.77

 vested

-0.70

SourceFile

-0.69

eln

-0.67

ola

-0.67

Oracle

-0.65

 Pact

-0.64

 otherwise

-0.64

yright

-0.64

POSITIVE LOGITS

srf

0.78

CHO

0.72

CHAPTER

0.70

©¶æ

0.68

 satell

0.68

 referen

0.68

pace

0.67

outube

0.67

 CONFIG

0.66

NetMessage

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.