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

answered

-0.77

ortunate

-0.72

lim

-0.72

exempt

-0.71

undo

-0.70

obl

-0.70

oshenko

-0.68

eny

-0.68

miss

-0.67

anny

-0.67

POSITIVE LOGITS

 behavi

0.73

ship

0.64

 MacBook

0.63

 dysph

0.62

SPONSORED

0.61

 Thames

0.60

 captcha

0.59

 views

0.59

 BART

0.59

Balt

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.