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

 Huma

-0.68

 Cunningham

-0.68

 Conce

-0.67

 opio

-0.66

 scen

-0.66

 Rivera

-0.62

cms

-0.61

sew

-0.61

pedia

-0.60

Zak

-0.60

POSITIVE LOGITS

ocene

0.79

sett

0.70

dropping

0.70

ersion

0.68

ffic

0.67

gency

0.67

venture

0.66

bre

0.63

 srfAttach

0.63

izards

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.