INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ngth

-0.75

gel

-0.70

mma

-0.68

idth

-0.67

Availability

-0.65

requisites

-0.65

fail

-0.64

haar

-0.64

 Kamp

-0.63

phant

-0.63

POSITIVE LOGITS

 Releases

0.70

 Neon

0.70

 Mich

0.68

 EVENTS

0.67

 fortune

0.67

 Emmy

0.67

Riy

0.65

 IMAGES

0.64

 Friendship

0.62

 penny

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.