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

RB

-0.75

çīĪ

-0.70

 Bunny

-0.68

Chr

-0.67

CN

-0.66

 chasing

-0.64

 tails

-0.63

WS

-0.62

RT

-0.62

 Rudolph

-0.61

POSITIVE LOGITS

Recommend

0.76

credit

0.73

rooms

0.69

graph

0.68

ãĤ¢ãĥ«

0.67

retty

0.67

ographies

0.66

interest

0.65

criminal

0.65

ettlement

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.