INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 setback

-0.81

 bully

-0.76

 defect

-0.74

 vend

-0.74

 reneg

-0.68

 grievances

-0.68

 sink

-0.68

 renegoti

-0.66

 lapse

-0.65

 Tough

-0.65

POSITIVE LOGITS

uras

0.82

 Cosponsors

0.76

 Canal

0.74

 Bris

0.73

HY

0.71

APS

0.71

apego

0.69

azel

0.68

hatt

0.67

ATS

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.