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

UTC

-0.69

ailability

-0.69

 Sawyer

-0.69

isible

-0.68

ļéĨĴ

-0.67

Ĭ

-0.66

gal

-0.64

 Holden

-0.64

 Katz

-0.63

ĨĴ

-0.62

POSITIVE LOGITS

yip

0.85

sylv

0.80

ribute

0.74

elist

0.74

onian

0.73

 sweat

0.69

erd

0.69

ewater

0.67

eret

0.67

ributes

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.