INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ħĭ

-0.72

²¾

-0.71

uthor

-0.68

cember

-0.67

ISSION

-0.66

 Vikings

-0.64

ikers

-0.64

 trainers

-0.64

Ĥ

-0.63

 patrols

-0.63

POSITIVE LOGITS

abet

0.73

Reviewed

0.72

inates

0.71

Emily

0.70

hetically

0.69

McA

0.68

âĨ

0.67

inia

0.66

 FIRE

0.65

aneously

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.