INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ãĥĵ

-0.71

ãĤ´

-0.69

isu

-0.67

 blasting

-0.67

ICT

-0.65

 docking

-0.65

Â¯Â¯Â¯Â¯

-0.63

OSS

-0.63

Þ

-0.62

ãĥ£

-0.61

POSITIVE LOGITS

etheless

0.69

than

0.68

gypt

0.67

 cous

0.67

].

0.64

 Democr

0.64

 Flavoring

0.64

ylum

0.62

hover

0.61

modules

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.