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

velength

-0.73

kj

-0.71

venue

-0.67

pas

-0.67

 deaf

-0.67

 subpoen

-0.67

racuse

-0.65

ouver

-0.62

rawdownloadcloneembedreportprint

-0.62

 earthqu

-0.62

POSITIVE LOGITS

ĺ

1.59

ı

0.74

isoft

0.74

 Rudd

0.72

alo

0.69

rov

0.66

-|

0.65

ART

0.64

STON

0.64

 deter

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.