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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istani

-0.75

 Creep

-0.75

emetery

-0.72

onom

-0.71

oÄŁ

-0.71

eria

-0.71

 Rican

-0.70

choice

-0.67

nom

-0.64

onomic

-0.63

POSITIVE LOGITS

Ware

0.71

 Flavoring

0.69

cial

0.66

irit

0.65

 render

0.64

ARM

0.63

UA

0.62

picture

0.62

EW

0.60

NK

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.