INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

orah

-0.75

aeus

-0.75

={

-0.71

pps

-0.65

elia

-0.65

MQ

-0.64

Ã³n

-0.63

onite

-0.63

aran

-0.63

mobi

-0.62

POSITIVE LOGITS

theless

0.74

 ancest

0.70

ly

0.68

 sensitivity

0.67

 Sabha

0.65

 safety

0.65

 sidx

0.65

 Enhancement

0.64

 turb

0.64

usting

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.