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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Wilderness

-0.69

 Cycl

-0.68

 Changing

-0.67

 depended

-0.64

 regenerate

-0.63

 depend

-0.63

pace

-0.62

 overw

-0.61

 cycl

-0.60

 Caucasus

-0.60

POSITIVE LOGITS

UME

0.76

AMI

0.71

wiki

0.71

lins

0.68

die

0.68

ELL

0.67

ANN

0.67

=>

0.65

OTT

0.64

ADRA

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.