INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

cffff

-0.78

emn

-0.76

eele

-0.76

omore

-0.76

rill

-0.75

estones

-0.75

hement

-0.74

esta

-0.74

ospons

-0.73

legate

-0.72

POSITIVE LOGITS

Gab

0.72

 Marx

0.68

 Audi

0.67

 antibiotic

0.67

NOW

0.65

SPONSORED

0.62

 Crusade

0.61

 Vand

0.61

Marx

0.60

 ..........

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.