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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orsche

-0.72

EO

-0.70

mingham

-0.69

leading

-0.65

orman

-0.64

 newcom

-0.64

 successor

-0.62

reditary

-0.61

 estab

-0.61

uesday

-0.61

POSITIVE LOGITS

ufact

0.77

 Phar

0.74

amples

0.74

gress

0.73

Interstitial

0.71

acters

0.71

igans

0.69

aband

0.66

Ô

0.66

Pers

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.