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

ayan

-0.90

Gaming

-0.86

yrinth

-0.77

uart

-0.76

Developer

-0.76

raltar

-0.76

aiman

-0.75

OIL

-0.75

ogg

-0.75

overe

-0.73

POSITIVE LOGITS

 resume

0.70

 Redux

0.70

nces

0.64

 rÃ©

0.64

 Commun

0.64

 Result

0.62

 Jama

0.61

ĩ

0.60

 Chart

0.60

CS

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.