INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

izophren

-0.81

kees

-0.76

chedel

-0.74

 Meditation

-0.73

netflix

-0.70

ouf

-0.69

pac

-0.68

fat

-0.67

onite

-0.67

 Awakens

-0.67

POSITIVE LOGITS

 Nanto

0.76

Suz

0.70

Cly

0.69

 Cerberus

0.64

 Becky

0.64

 Seraph

0.63

 Diane

0.62

 Leopard

0.62

 Leigh

0.62

Meet

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.