INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Sanders

-0.61

 Lewis

-0.61

 Ital

-0.59

 Richardson

-0.59

 breakdown

-0.59

 Females

-0.59

Adjust

-0.59

uin

-0.58

ĪĴ

-0.58

 Abrams

-0.58

POSITIVE LOGITS

attach

0.75

company

0.70

pair

0.69

ãĥ¢

0.66

æ©Ł

0.65

 moth

0.65

 pomp

0.65

ichick

0.64

number

0.64

 Devil

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.