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

 tremend

-0.86

 spor

-0.74

 bicy

-0.73

drivers

-0.69

 carrot

-0.68

inately

-0.67

 mosqu

-0.67

 lapt

-0.64

coh

-0.64

 Scion

-0.63

POSITIVE LOGITS

ãĥĨ

0.78

alky

0.72

angelo

0.69

 DirectX

0.68

anka

0.65

olds

0.65

ichick

0.64

 Cosponsors

0.64

================================================================

0.64

ibrary

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.