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

Ä

-0.69

 pandemonium

-0.65

 Showdown

-0.64

uador

-0.64

izoph

-0.63

 opinion

-0.63

 incons

-0.62

 lawsuit

-0.61

 uproar

-0.60

 Commissioners

-0.60

POSITIVE LOGITS

stellar

0.86

ledged

0.76

rentice

0.74

SQL

0.73

hin

0.73

elia

0.72

ILA

0.72

utsche

0.70

iron

0.70

adi

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.