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

gage

-0.71

sites

-0.69

 Logged

-0.66

bed

-0.65

 shoes

-0.63

 curfew

-0.62

0000000000000000

-0.62

isa

-0.61

 beds

-0.60

 babys

-0.60

POSITIVE LOGITS

ngth

0.75

else

0.68

 GDDR

0.68

Ã§a

0.64

icz

0.63

âĹ¼

0.62

tions

0.61

isans

0.61

CPC

0.60

Boe

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.