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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

horn

-0.72

 Forth

-0.67

ENA

-0.63

 Bulgar

-0.62

Ã»

-0.62

Stone

-0.62

 Influence

-0.61

ATED

-0.61

 Canary

-0.60

 DRAGON

-0.59

POSITIVE LOGITS

 cleaner

0.65

hett

0.65

natureconservancy

0.62

COL

0.62

igi

0.61

 scrub

0.61

 downloadable

0.60

 inclusive

0.59

ettlement

0.59

appy

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.