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

ibaba

-0.80

igil

-0.70

igo

-0.70

undle

-0.70

irm

-0.67

pack

-0.67

kr

-0.67

packs

-0.66

 redes

-0.65

ense

-0.63

POSITIVE LOGITS

 stunts

0.72

onel

0.66

ITIES

0.64

 Goldberg

0.64

CLASSIFIED

0.64

America

0.62

 Godd

0.61

CHAPTER

0.61

 Thatcher

0.60

 sterling

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.