INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Accessory

-0.76

leness

-0.69

Afee

-0.68

Â·Â·

-0.66

Frag

-0.65

hy

-0.64

attribute

-0.64

Ward

-0.64

pering

-0.64

 Wraith

-0.64

POSITIVE LOGITS

esar

0.81

ancial

0.79

aced

0.74

ARDS

0.74

awks

0.72

wcs

0.70

tarian

0.67

 trusts

0.67

agle

0.65

iated

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.