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

Reviewer

-0.75

lio

-0.73

 Myanmar

-0.67

externalActionCode

-0.67

 Meridian

-0.66

enary

-0.66

ceed

-0.65

 Rohingya

-0.65

ÃĽÃĽ

-0.64

quished

-0.64

POSITIVE LOGITS

 belts

0.80

peak

0.79

BILITIES

0.72

 mustache

0.68

 channels

0.68

ench

0.67

gas

0.66

 warr

0.64

 sock

0.64

 belt

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.