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

ãĤ´ãĥ³

-0.72

 couch

-0.62

 ogre

-0.62

 transcription

-0.62

IG

-0.61

æĸ

-0.61

 crank

-0.61

 istg

-0.60

ä»£

-0.60

 accuser

-0.60

POSITIVE LOGITS

ienne

0.81

uador

0.79

paio

0.77

berus

0.77

ocene

0.77

LECT

0.75

cks

0.75

Ã§a

0.74

Buff

0.74

cery

0.73

Activations Density 0.000%

No Known Activations

This feature has no known activations.