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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Krish

-0.64

omo

-0.64

yss

-0.63

Por

-0.63

Yen

-0.63

 Cathy

-0.62

 Braun

-0.62

 appellant

-0.62

 comparative

-0.61

 Fowler

-0.61

POSITIVE LOGITS

ombies

0.79

Tickets

0.73

FIG

0.73

Stars

0.72

STEM

0.72

FU

0.71

USD

0.71

RH

0.71

OTS

0.70

itans

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.