INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

agonist

-0.89

dullah

-0.73

jriwal

-0.70

amous

-0.67

taker

-0.65

ayson

-0.64

gat

-0.63

gger

-0.63

din

-0.63

culosis

-0.62

POSITIVE LOGITS

 Decoder

0.68

merce

0.65

OND

0.64

 FANTASY

0.64

MAT

0.61

 COURT

0.61

 Represent

0.61

verages

0.61

 FORM

0.61

 Designs

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.