INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Ambrose

-0.75

 Creed

-0.75

venants

-0.72

McF

-0.71

unction

-0.66

 eleph

-0.64

essen

-0.63

uls

-0.62

ptions

-0.62

lication

-0.61

POSITIVE LOGITS

kees

0.78

cffffcc

0.77

Plot

0.71

rama

0.69

Fake

0.66

Spread

0.65

bum

0.64

iframe

0.64

avid

0.63

atar

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.