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

rek

-0.75

 Spiegel

-0.74

esa

-0.73

nel

-0.68

Trigger

-0.67

 spit

-0.67

ington

-0.66

pelled

-0.63

ike

-0.63

mit

-0.62

POSITIVE LOGITS

 Solitaire

0.87

 Tsukuyomi

0.76

 Skydragon

0.74

Yo

0.70

æ©Ł

0.69

çīĪ

0.66

$$$$

0.64

Reloaded

0.64

asters

0.62

 killer

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.