INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

senal

-0.89

 Canaver

-0.77

staking

-0.73

-0.68

mone

-0.68

pse

-0.66

CG

-0.65

 arrang

-0.65

 stroke

-0.65

redo

-0.65

POSITIVE LOGITS

gee

0.73

fell

0.72

eries

0.69

ards

0.69

orius

0.68

ãĢ

0.68

yond

0.67

aum

0.65

 Raspberry

0.64

Bom

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.