INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

NetMessage

-0.78

ideshow

-0.77

Storage

-0.76

osures

-0.76

dies

-0.72

odore

-0.70

auri

-0.69

aspers

-0.68

ŃĶ

-0.67

rand

-0.65

POSITIVE LOGITS

PM

0.68

 Freak

0.67

 Sheen

0.66

ricular

0.66

essional

0.61

 Evangel

0.61

ioxide

0.61

 Course

0.61

 BLACK

0.59

 course

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.