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

otom

-0.82

onga

-0.78

iets

-0.77

iband

-0.72

tesy

-0.72

oha

-0.72

ikh

-0.71

 sidel

-0.70

onia

-0.70

 induct

-0.68

POSITIVE LOGITS

dll

0.81

0000000000000000

0.81

 Afterwards

0.79

Comments

0.78

 They

0.76

DragonMagazine

0.74

aspx

0.73

Links

0.72

Management

0.71

 Then

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.