INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ike

-0.71

Mub

-0.71

itled

-0.70

Downloadha

-0.67

ibaba

-0.67

sha

-0.66

jab

-0.65

 Hassan

-0.64

ha

-0.64

rek

-0.63

POSITIVE LOGITS

âĦ¢:

0.81

atar

0.71

asions

0.69

 Seconds

0.67

estern

0.67

aple

0.64

î

0.63

 Souls

0.62

 Span

0.61

 Scal

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.