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

BILITIES

-0.84

 Version

-0.72

 lett

-0.70

Version

-0.67

BILITY

-0.66

 Transparency

-0.66

Choice

-0.64

 Dare

-0.63

Provider

-0.62

 Entry

-0.62

POSITIVE LOGITS

miah

0.74

earchers

0.73

 sqor

0.69

 captcha

0.67

é¾

0.67

ihu

0.66

 exting

0.66

clinton

0.66

icularly

0.65

sem

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.