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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rep

-0.72

inately

-0.69

soType

-0.69

ploy

-0.66

ezvous

-0.65

escal

-0.65

CTR

-0.65

swer

-0.65

quer

-0.64

owsky

-0.62

POSITIVE LOGITS

ħĭ

0.86

ESCO

0.77

 Flavoring

0.77

oult

0.71

natureconservancy

0.71

Sequ

0.70

 Eternity

0.66

 Swords

0.63

 nonex

0.62

Reviewer

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.