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

Rio

-0.71

 Guan

-0.65

Sod

-0.65

WTC

-0.65

Nat

-0.62

 Sports

-0.61

Dam

-0.61

 Canal

-0.60

otle

-0.60

 Garry

-0.60

POSITIVE LOGITS

 LIKE

0.89

 like

0.83

catentry

0.73

enegger

0.73

\\\\\\\\

0.72

enfranch

0.72

 Logged

0.69

iversal

0.69

Like

0.67

orthy

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.