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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

emale

-0.73

artist

-0.66

emon

-0.65

luster

-0.65

é¾įå¥ĳå£«

-0.64

rett

-0.63

iov

-0.63

STD

-0.62

occ

-0.61

alking

-0.61

POSITIVE LOGITS

hower

0.79

 Demand

0.74

puted

0.68

selves

0.64

izer

0.61

demand

0.61

UP

0.61

ts

0.60

 Doctors

0.59

 Vive

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.