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

isoft

-0.75

 unaff

-0.68

 embargo

-0.66

 withdrawn

-0.65

 Anchorage

-0.64

 hostages

-0.64

 Oslo

-0.63

iliated

-0.63

 poaching

-0.63

 Bulg

-0.62

POSITIVE LOGITS

)]

0.73

Writ

0.71

Bra

0.70

Cap

0.69

MSN

0.68

dll

0.68

ãĥ£

0.67

notation

0.66

edy

0.63

Mo

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.