INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

matter

-0.67

course

-0.63

olves

-0.63

helle

-0.63

 progression

-0.61

 relegation

-0.61

 balance

-0.60

 Squirrel

-0.60

gow

-0.60

DCS

-0.58

POSITIVE LOGITS

renheit

0.97

osate

0.91

 Arabia

0.77

heid

0.73

 Aram

0.69

ItemTracker

0.68

arta

0.67

SPONSORED

0.65

acus

0.65

 Salman

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.