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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Meadows

-0.79

 Incarn

-0.76

 Disorder

-0.74

 Fever

-0.73

dullah

-0.72

ocene

-0.71

 Chow

-0.71

 Enhancement

-0.69

ãĤĬ

-0.69

 Diseases

-0.68

POSITIVE LOGITS

 testers

0.76

urally

0.74

 glim

0.72

yright

0.72

 tabs

0.70

 sampling

0.70

astically

0.69

 workings

0.67

inately

0.67

 branching

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.