INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Markus

-0.82

kowski

-0.80

 Berks

-0.73

ameda

-0.70

agers

-0.69

wikipedia

-0.69

iencies

-0.68

Ole

-0.67

 Neal

-0.67

Compan

-0.66

POSITIVE LOGITS

 underworld

0.74

 terminating

0.68

 lodge

0.65

cellence

0.64

 weeping

0.64

ngth

0.64

withstanding

0.63

PO

0.63

ilaterally

0.62

 masse

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.