INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Leilan

-0.88

emon

-0.85

;;;;;;;;;;;;

-0.67

itude

-0.67

 Cron

-0.66

 arte

-0.66

Dat

-0.65

å¥³

-0.65

 Siren

-0.64

Yor

-0.63

POSITIVE LOGITS

inals

0.77

ĪĴ

0.75

lished

0.74

velt

0.72

VIDIA

0.71

inally

0.70

yrinth

0.68

VIEW

0.66

rious

0.66

original

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.