INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

wards

-0.78

rient

-0.75

ciplinary

-0.74

 outlet

-0.67

 environment

-0.65

tal

-0.65

team

-0.64

 bikini

-0.64

 background

-0.63

 options

-0.63

POSITIVE LOGITS

 Presbyter

0.87

DK

0.80

kefeller

0.76

 Pengu

0.73

odox

0.73

arry

0.70

 Lutheran

0.68

NK

0.68

 Convention

0.68

ħĭ

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.