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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Canberra

-0.69

ModLoader

-0.69

REL

-0.67

velength

-0.67

gig

-0.66

RTX

-0.64

 fastball

-0.64

 Lesbian

-0.64

 Kushner

-0.63

 Epstein

-0.62

POSITIVE LOGITS

artifacts

0.92

conom

0.80

utch

0.79

aks

0.75

nov

0.71

progress

0.68

;;;;;;;;;;;;

0.68

},"

0.65

humane

0.65

axter

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.