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

 slang

-0.70

 glac

-0.69

 meteor

-0.68

 scrape

-0.68

 graffiti

-0.67

 snipers

-0.66

 souven

-0.64

 geography

-0.62

 subdiv

-0.62

 Tire

-0.61

POSITIVE LOGITS

aii

0.89

ipolar

0.84

efe

0.77

ELF

0.77

ichick

0.76

anon

0.75

efer

0.75

HER

0.75

oldemort

0.75

uclear

0.74

Activations Density 0.000%

No Known Activations

This feature has no known activations.