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

ELY

-0.68

 Strat

-0.68

 Berman

-0.68

Rim

-0.68

 Refuge

-0.68

Gig

-0.66

andi

-0.65

 Butt

-0.64

ollar

-0.64

 Advertisement

-0.64

POSITIVE LOGITS

 reluct

0.75

é¾įå¥ĳå£«

0.74

unknown

0.73

cffff

0.70

 unfocusedRange

0.70

 refreshed

0.69

rex

0.69

+#

0.68

ģĸ

0.67

 ashore

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.