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

ĸļ

-0.99

iatus

-0.81

 Caption

-0.78

itivity

-0.77

loo

-0.77

ģ«

-0.76

izont

-0.76

Ĥİ

-0.76

Ëľ

-0.75

ashtra

-0.74

POSITIVE LOGITS

 bearer

0.72

 postage

0.68

aches

0.68

 tower

0.67

 pier

0.65

 nurt

0.65

PI

0.64

 pillar

0.63

 scholar

0.63

 wards

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.