INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Notting

-0.77

itri

-0.75

 istg

-0.71

 scares

-0.66

 Sith

-0.65

 dstg

-0.61

 Strateg

-0.61

ologue

-0.60

idding

-0.60

 Poker

-0.60

POSITIVE LOGITS

swing

0.78

DonaldTrump

0.75

floor

0.69

quartered

0.68

GROUND

0.67

apsed

0.65

itialized

0.64

Draft

0.63

opped

0.62

WIND

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.