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

phrine

-0.75

 Bowman

-0.73

 Masquerade

-0.69

Jet

-0.67

 Dmit

-0.67

icter

-0.65

Accessory

-0.65

sym

-0.64

 Freeman

-0.64

 Thumbnails

-0.63

POSITIVE LOGITS

ikk

0.80

ispers

0.68

so

0.66

 suscept

0.65

 acron

0.64

ocry

0.63

ancies

0.62

 jurisdictions

0.61

raltar

0.61

 Rated

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.