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

fold

-0.70

mpeg

-0.69

lot

-0.69

bury

-0.68

lift

-0.68

CF

-0.68

 Maver

-0.67

sic

-0.67

Oracle

-0.66

QB

-0.66

POSITIVE LOGITS

 probe

0.69

waukee

0.69

imony

0.65

 ultrasound

0.64

osponsors

0.62

irtual

0.61

 reference

0.59

 microwave

0.59

illin

0.59

 epigen

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.