INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

DAQ

-0.74

WARD

-0.70

 Marse

-0.68

OTA

-0.67

 Rica

-0.66

 Bret

-0.66

CLOSE

-0.66

âķĲ

-0.63

pty

-0.63

HAHA

-0.62

POSITIVE LOGITS

deen

0.95

lihood

0.81

 pier

0.73

 engagement

0.69

wegian

0.69

skirts

0.64

 Virtue

0.63

chest

0.62

ateg

0.62

 Territories

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.