INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

rique

-0.67

 Guerrero

-0.66

opathy

-0.64

icism

-0.63

Ã°

-0.62

ORT

-0.62

 Olson

-0.62

 fitt

-0.62

å¿

-0.61

 Foley

-0.61

POSITIVE LOGITS

bart

0.79

 Archdemon

0.74

Dragon

0.71

 intel

0.70

mega

0.69

*/(

0.66

elsius

0.64

 Companion

0.64

Brave

0.63

 Cheong

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.