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

adra

-0.75

ocaust

-0.72

ocene

-0.72

 Gardens

-0.71

utral

-0.68

appiness

-0.68

ENC

-0.66

aria

-0.65

ettle

-0.65

uba

-0.63

POSITIVE LOGITS

 Dull

0.69

 tact

0.66

 understatement

0.64

 etiquette

0.63

ifully

0.63

 inertia

0.63

 Beir

0.62

 Stain

0.62

 talent

0.61

inally

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.