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

Times

-0.73

 rall

-0.70

ixt

-0.70

-0.68

 Tomorrow

-0.67

 Citation

-0.67

 thereto

-0.66

 Feet

-0.65

Ft

-0.65

 Andrews

-0.64

POSITIVE LOGITS

iery

0.74

cling

0.73

 potent

0.67

gently

0.67

flush

0.64

helial

0.64

 inflammation

0.63

 cleansing

0.63

cle

0.63

 cycl

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.