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

 Alam

-0.70

 Prin

-0.65

 cabinets

-0.64

Spain

-0.63

 holders

-0.63

 Britann

-0.63

Alm

-0.62

bars

-0.62

Benz

-0.62

holders

-0.61

POSITIVE LOGITS

icum

0.86

erous

0.77

itars

0.75

 redes

0.73

 facult

0.72

sed

0.72

itton

0.70

culus

0.69

 appropriation

0.69

orns

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.