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

 itch

-0.74

ufact

-0.68

thren

-0.63

 Qing

-0.62

 fixme

-0.62

 Cheong

-0.61

antle

-0.61

nesota

-0.60

 Phant

-0.59

Nou

-0.59

POSITIVE LOGITS

pert

2.31

atro

0.85

aminer

0.83

utenberg

0.77

Reviewed

0.77

brid

0.72

lishes

0.68

rence

0.67

pees

0.67

anos

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.