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

imet

-0.78

 wards

-0.70

Xia

-0.69

 crypt

-0.68

Ich

-0.67

Bei

-0.65

peer

-0.63

staking

-0.63

hement

-0.62

 Xiang

-0.61

POSITIVE LOGITS

advertising

0.78

assad

0.74

yss

0.74

cv

0.73

UAL

0.72

nv

0.71

acted

0.71

ilde

0.71

uded

0.69

ACTION

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.