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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 corrid

-0.92

 mathemat

-0.80

 Chero

-0.70

KNOWN

-0.69

uca

-0.69

Ô

-0.67

Īè

-0.66

eatures

-0.66

 Palest

-0.62

 predec

-0.62

POSITIVE LOGITS

racuse

0.72

 Tsukuyomi

0.70

isky

0.69

ubi

0.69

itbart

0.69

plate

0.67

heit

0.66

 Meanwhile

0.64

heet

0.64

orsi

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.