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

oute

-0.71

 contrace

-0.70

BG

-0.66

 Stephens

-0.64

inates

-0.63

-0.62

 Baghd

-0.60

 cooker

-0.60

 Skinner

-0.59

Hunt

-0.59

POSITIVE LOGITS

MENTS

0.72

oken

0.69

å£«

0.68

rics

0.66

aldo

0.65

vous

0.64

ecd

0.64

misc

0.63

 Janeiro

0.63

orth

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.