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

owship

-0.76

conservancy

-0.74

itect

-0.73

orage

-0.72

ogether

-0.70

itten

-0.68

nance

-0.67

 Citiz

-0.67

atform

-0.65

 Flavoring

-0.63

POSITIVE LOGITS

 Guerrero

0.75

elo

0.72

Sop

0.66

chrom

0.64

Mus

0.64

Chat

0.64

oldemort

0.64

Slot

0.63

Pick

0.63

Kid

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.