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

abad

-0.75

Aj

-0.73

Īè

-0.70

 evacuation

-0.65

Downloadha

-0.65

 Opening

-0.65

 advertis

-0.63

 premises

-0.62

saf

-0.62

CRIP

-0.61

POSITIVE LOGITS

perty

0.78

sis

0.70

esta

0.68

amac

0.67

lip

0.67

sle

0.62

pite

0.61

lich

0.60

 stagn

0.60

kefeller

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.