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

ĳ

-0.76

 Norwich

-0.69

 galleries

-0.69

Hav

-0.69

ests

-0.67

 Trafford

-0.67

 Ghostbusters

-0.66

 Cologne

-0.66

 Tanz

-0.65

 Wembley

-0.64

POSITIVE LOGITS

.):

0.68

âĢ¦)

0.65

 Attribution

0.63

 hereby

0.62

nery

0.61

 affiliate

0.60

ureau

0.60

âĢ¦]

0.60

unal

0.59

 pine

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.