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

 Savior

-0.76

agall

-0.68

 Sacrifice

-0.67

 Init

-0.66

idelines

-0.66

 Hell

-0.66

 Curse

-0.65

 Hannibal

-0.64

ESV

-0.63

 Civilization

-0.63

POSITIVE LOGITS

 ........

0.79

ensable

0.78

partisan

0.67

ensed

0.65

ngth

0.64

oned

0.62

perty

0.62

essen

0.61

otypes

0.60

phony

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.