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

soever

-0.73

 race

-0.71

 utilization

-0.69

ģĸ

-0.67

iotic

-0.66

 institution

-0.66

ciating

-0.65

angular

-0.64

 organism

-0.64

onian

-0.62

POSITIVE LOGITS

 Â¯

0.71

 Beir

0.70

Za

0.69

jam

0.67

jong

0.65

asters

0.65

ultimate

0.65

sk

0.65

tails

0.64

 Siem

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.