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

antry

-0.77

glas

-0.75

croft

-0.75

lance

-0.74

ienne

-0.72

ornia

-0.72

semble

-0.71

icle

-0.70

aney

-0.69

cli

-0.69

POSITIVE LOGITS

 misunder

0.67

aution

0.67

ãĤ¤ãĥĪ

0.66

ammy

0.63

POV

0.60

 orgasm

0.59

 swayed

0.59

faced

0.59

DragonMagazine

0.58

 Honor

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.