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

 hopping

-0.81

 beads

-0.76

hop

-0.70

 lobe

-0.68

 heading

-0.67

fro

-0.65

vic

-0.64

aft

-0.64

 hops

-0.62

addle

-0.62

POSITIVE LOGITS

 Flavoring

0.82

 Helpful

0.80

ufact

0.78

anke

0.75

Privacy

0.73

Show

0.72

icio

0.72

anth

0.70

iability

0.69

Story

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.