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

DNA

-0.77

fact

-0.73

ickets

-0.69

illard

-0.69

facts

-0.69

utral

-0.69

Mit

-0.66

 Rudy

-0.66

ology

-0.66

 toast

-0.65

POSITIVE LOGITS

 newcom

0.69

raq

0.69

Ake

0.65

ellow

0.65

ram

0.62

 Byrne

0.62

ahime

0.62

 Awakens

0.61

MAS

0.61

 writ

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.