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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SET

-0.74

AZ

-0.67

TOP

-0.66

SU

-0.66

Utah

-0.65

ENCE

-0.65

ivas

-0.65

UTE

-0.64

APTER

-0.64

amber

-0.63

POSITIVE LOGITS

berries

0.71

 Calais

0.71

 Lans

0.67

ebook

0.64

Mine

0.64

ssh

0.63

 Westminster

0.61

Trust

0.61

esan

0.60

ngth

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.