INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

(-

-0.74

(+

-0.72

icious

-0.71

(-

-0.71

ific

-0.69

asma

-0.67

BN

-0.64

 decl

-0.64

ides

-0.64

Ö¼

-0.61

POSITIVE LOGITS

terness

0.82

sole

0.67

 sqor

0.65

 Volunteer

0.63

 Takeru

0.62

 Mehran

0.62

querque

0.61

nep

0.60

phot

0.60

Yue

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.