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

 Marketable

-0.71

ovic

-0.69

about

-0.65

forge

-0.63

Downloadha

-0.62

pton

-0.61

 Required

-0.59

 Chambers

-0.59

 Alexa

-0.59

sight

-0.58

POSITIVE LOGITS

enhagen

0.80

ouver

0.77

achus

0.72

ributes

0.66

ateurs

0.66

illac

0.66

ONSORED

0.66

rance

0.65

rative

0.65

ande

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.