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

 Monstrous

-0.82

rating

-0.68

 Lump

-0.64

 flames

-0.63

bp

-0.63

%]

-0.61

erker

-0.61

killer

-0.61

 Nichols

-0.61

glers

-0.60

POSITIVE LOGITS

essions

0.71

ulz

0.70

 SHOW

0.66

all

0.65

 TRANS

0.63

 Fargo

0.62

Dial

0.62

Ã¼

0.61

Interstitial

0.61

 Yose

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.