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

 visionary

-0.71

etheless

-0.70

 Berm

-0.68

oola

-0.65

 sage

-0.65

 municip

-0.63

 valiant

-0.63

 prudent

-0.63

ctory

-0.63

 Habit

-0.62

POSITIVE LOGITS

OH

0.78

ape

0.75

ARS

0.74

ONES

0.67

OO

0.66

meat

0.66

APE

0.65

Reply

0.64

rin

0.63

IO

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.