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

©¶æ¥µ

-0.92

ĸļ

-0.89

 satell

-0.75

¿½

-0.75

Īè

-0.75

anchester

-0.70

antha

-0.69

guyen

-0.68

ð

-0.68

 Slay

-0.66

POSITIVE LOGITS

oy

0.75

gard

0.73

 Sovereign

0.72

 Pact

0.70

etus

0.68

 Columb

0.67

uble

0.66

rog

0.66

eli

0.66

ta

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.