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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

intage

-0.74

levard

-0.68

itton

-0.67

 indie

-0.65

rent

-0.65

 conflic

-0.64

soType

-0.64

uj

-0.64

aturdays

-0.64

 Shopping

-0.63

POSITIVE LOGITS

chlor

0.76

anon

0.70

 Aristotle

0.68

throp

0.66

username

0.64

 Herod

0.63

DCS

0.63

thia

0.61

 wonders

0.61

nesota

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.