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

ween

-0.81

 resent

-0.72

AFTA

-0.67

osta

-0.65

 despise

-0.64

²¾

-0.64

geist

-0.63

 olive

-0.62

isol

-0.62

helle

-0.62

POSITIVE LOGITS

chenko

0.74

 Viktor

0.73

ovsky

0.66

 Entered

0.65

 Miko

0.65

anova

0.64

 Kard

0.63

 Kend

0.63

 Wend

0.62

 Stefan

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.