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

£ı

-0.81

 Patriarch

-0.75

itary

-0.75

 unfocusedRange

-0.70

 Squirrel

-0.67

uyomi

-0.66

advertisement

-0.66

Ĭ

-0.66

aghd

-0.65

Īè

-0.65

POSITIVE LOGITS

abouts

0.74

hops

0.68

 Johannes

0.67

redo

0.67

rica

0.67

hop

0.67

boot

0.65

ROM

0.65

Â£

0.64

berto

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.