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

agar

-0.84

boss

-0.76

sburg

-0.73

hold

-0.72

rog

-0.69

oken

-0.67

ached

-0.67

ell

-0.67

gg

-0.66

aff

-0.66

POSITIVE LOGITS

terday

0.71

 cryst

0.70

 scrut

0.70

 foss

0.69

 incent

0.68

 livest

0.68

rpm

0.67

oult

0.67

 ultr

0.66

 Sask

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.