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

 shit

-0.76

aroo

-0.73

 antic

-0.71

 Shit

-0.68

 swast

-0.68

 Donetsk

-0.67

 fres

-0.66

 insensitive

-0.65

letter

-0.65

 ****

-0.64

POSITIVE LOGITS

JS

0.84

eq

0.80

PE

0.78

abeth

0.76

NL

0.76

ESH

0.75

eph

0.74

VERT

0.73

PU

0.73

 Noir

0.73

Activations Density 0.000%

No Known Activations

This feature has no known activations.