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

lace

-0.80

rehend

-0.74

[|

-0.74

yip

-0.73

 charact

-0.70

erity

-0.70

heet

-0.68

robat

-0.67

oother

-0.67

dayName

-0.66

POSITIVE LOGITS

 Editors

0.79

inka

0.74

Ca

0.73

 Gott

0.72

 Bernstein

0.69

 Chern

0.69

 registry

0.69

 Feldman

0.66

 Gors

0.65

 Bolton

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.