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

 Mush

-0.72

 breath

-0.69

 scratch

-0.66

 spect

-0.63

 abort

-0.61

 secondary

-0.59

 towel

-0.59

 Terra

-0.59

 Niet

-0.58

 taste

-0.58

POSITIVE LOGITS

 Says

0.87

Wars

0.73

iser

0.70

fram

0.70

elling

0.70

wagen

0.69

imedia

0.69

olphins

0.68

 said

0.68

Daily

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.