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

Negative Logits

apo

-0.76

idelines

-0.73

ationally

-0.71

ensation

-0.71

opsy

-0.71

aldehyde

-0.70

eda

-0.70

iate

-0.69

ops

-0.69

anchester

-0.68

POSITIVE LOGITS

 Christensen

0.75

 Petersen

0.72

 Leone

0.71

 Apost

0.68

pher

0.68

Host

0.67

script

0.65

 Hubbard

0.64

wed

0.64

 Stevens

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.