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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

agame

-0.78

itud

-0.74

Kick

-0.73

ategory

-0.66

icio

-0.65

raid

-0.65

ardy

-0.65

redd

-0.65

asso

-0.63

 enthusi

-0.62

POSITIVE LOGITS

*/(

0.76

 Marketplace

0.67

 Portal

0.67

 Sins

0.61

 Gaia

0.60

 Scorp

0.59

 midst

0.58

Sphere

0.58

 cake

0.58

 atmosphere

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.