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

uca

-0.83

-0.77

 Associates

-0.76

Pg

-0.73

ACA

-0.73

AMA

-0.72

AF

-0.71

chester

-0.69

afe

-0.69

Ã§a

-0.69

POSITIVE LOGITS

 fulfillment

0.77

 buyer

0.76

 harvest

0.76

 fifth

0.75

 Torah

0.75

 burden

0.75

 whole

0.74

 subdivision

0.72

 culmination

0.70

 blessing

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.