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

sung

-0.76

 McCoy

-0.73

ebted

-0.73

%]

-0.71

ÂłÂłÂłÂł

-0.70

VERTISEMENT

-0.70

CPU

-0.69

âĸº

-0.68

CLASSIFIED

-0.66

nyder

-0.66

POSITIVE LOGITS

iquette

0.73

 respecting

0.66

 Guides

0.64

jon

0.64

wyn

0.63

 ramps

0.61

raine

0.61

ridges

0.60

 besie

0.59

ivia

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.