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

SourceFile

-0.81

Haunted

-0.72

quickShipAvailable

-0.71

VID

-0.68

LV

-0.67

IGH

-0.66

BT

-0.65

Ko

-0.65

Alert

-0.62

URI

-0.62

POSITIVE LOGITS

rica

0.83

romeda

0.78

ptions

0.75

gements

0.73

senal

0.69

osure

0.69

orgetown

0.67

 Marketable

0.67

emort

0.67

hari

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.