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

¥µ

-0.84

Merit

-0.68

emo

-0.66

itte

-0.61

 Cohn

-0.61

 macros

-0.60

 Sapp

-0.58

alky

-0.58

ãģĦ

-0.56

itta

-0.56

POSITIVE LOGITS

ividual

0.72

levard

0.71

asonic

0.70

geries

0.70

phrine

0.68

 Laos

0.68

amine

0.67

 guiActiveUn

0.66

gment

0.66

assic

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.