INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ĸļ

-0.85

ushi

-0.84

adr

-0.84

uo

-0.84

anova

-0.84

UA

-0.78

agara

-0.78

keyes

-0.75

owa

-0.72

ika

-0.71

POSITIVE LOGITS

 pleas

0.70

nesday

0.68

 announcements

0.67

 guiActiveUn

0.64

 pans

0.63

 poles

0.62

 hind

0.62

 newsp

0.62

 Palest

0.61

 bulletin

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.