INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

channelAvailability

-0.87

ullivan

-0.64

 Operator

-0.62

rown

-0.62

 Nets

-0.61

 Bastard

-0.60

 Wond

-0.59

 Akin

-0.59

icularly

-0.58

DragonMagazine

-0.58

POSITIVE LOGITS

lation

0.73

atos

0.72

ethe

0.70

phal

0.70

âĻ

0.69

cases

0.68

bra

0.68

wer

0.67

eria

0.66

ule

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.