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

itals

-0.85

gut

-0.77

inery

-0.76

ently

-0.72

cade

-0.71

jured

-0.70

Sanders

-0.69

ip

-0.69

stru

-0.67

ital

-0.67

POSITIVE LOGITS

ciating

0.81

ħĭ

0.76

 Instant

0.73

 Downloads

0.73

 Portable

0.67

 Saying

0.64

ormal

0.64

odor

0.63

 Defin

0.62

SPL

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.