INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Ranked

-0.68

SHARE

-0.67

amer

-0.66

iol

-0.61

reon

-0.60

ness

-0.60

azon

-0.59

âĹ¼

-0.59

xtap

-0.59

sterdam

-0.58

POSITIVE LOGITS

any

1.22

ANY

1.05

 anybody

1.03

 anyone

0.99

 anything

0.99

anything

0.96

 anytime

0.92

any

0.86

Any

0.84

Any

0.79

Activations Density 0.000%

No Known Activations

This feature has no known activations.