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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 guiActiveUn

-0.67

ARDIS

-0.66

 visc

-0.63

 Minotaur

-0.60

ocative

-0.59

aeus

-0.58

 Manufacturer

-0.58

 reigning

-0.57

 birthplace

-0.57

 endogenous

-0.56

POSITIVE LOGITS

Las

0.73

adoes

0.68

ils

0.66

ciples

0.65

olson

0.65

uggets

0.63

Pac

0.63

iah

0.62

opsy

0.61

astics

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.