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

Accessory

-0.85

âĨ

-0.79

edge

-0.78

writ

-0.76

ought

-0.76

Spec

-0.75

stood

-0.74

ACTED

-0.74

effic

-0.73

received

-0.71

POSITIVE LOGITS

atown

0.64

 clipboard

0.63

quished

0.63

 languages

0.62

 Lumpur

0.61

 apologise

0.61

 counselling

0.61

sterdam

0.60

utory

0.60

 teens

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.