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

PsyNetMessage

-0.72

 Liberties

-0.71

Merit

-0.69

 Manifest

-0.68

advertisement

-0.68

body

-0.67

Interstitial

-0.66

lings

-0.65

utenberg

-0.64

ogens

-0.64

POSITIVE LOGITS

euro

0.73

RS

0.69

oute

0.64

DS

0.63

 reconc

0.60

yn

0.60

ocl

0.60

wagen

0.60

pec

0.60

iere

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.