INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

strings

-0.72

eenth

-0.70

 thirst

-0.66

 Rebellion

-0.65

 Crusade

-0.64

 repatri

-0.63

recomm

-0.62

aign

-0.61

 Triangle

-0.61

marriage

-0.60

POSITIVE LOGITS

essee

0.86

obyl

0.75

osher

0.74

ature

0.71

ATURES

0.70

Clock

0.68

atable

0.68

emis

0.67

ãĤ´

0.67

 srfAttach

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.