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

aven

-0.82

uggest

-0.80

States

-0.79

aturdays

-0.72

reprene

-0.70

ornings

-0.69

olute

-0.68

ahu

-0.66

enf

-0.66

grounds

-0.66

POSITIVE LOGITS

etc

0.75

 whiff

0.70

 Tyrann

0.67

 Kuro

0.64

 Lauder

0.64

LOL

0.64

 Osiris

0.63

eh

0.63

huh

0.62

sticks

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.