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

BSD

-0.80

olphin

-0.72

behavior

-0.70

ilib

-0.66

olester

-0.65

orc

-0.65

 Intervention

-0.65

 viol

-0.64

policy

-0.63

 emanc

-0.63

POSITIVE LOGITS

 Jagu

0.81

ÙĴ

0.74

aterasu

0.71

 HuffPost

0.67

 Mehran

0.67

eries

0.66

 Newcastle

0.65

packs

0.65

ãĥ¼ãĥ³

0.64

monton

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.