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

hall

-0.70

axter

-0.69

ieg

-0.67

arthy

-0.67

ieu

-0.62

words

-0.62

meric

-0.61

matically

-0.60

akura

-0.60

ordes

-0.60

POSITIVE LOGITS

caster

0.73

iPhone

0.70

BSD

0.67

erto

0.66

 targ

0.64

upt

0.64

warm

0.63

brew

0.63

handle

0.61

qt

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.