INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 challeng

-0.77

laus

-0.74

kee

-0.73

oth

-0.72

orah

-0.71

minist

-0.71

NetMessage

-0.70

east

-0.69

ties

-0.67

estone

-0.67

POSITIVE LOGITS

ISI

0.67

 attention

0.64

 inbox

0.64

AQ

0.62

 absor

0.61

omial

0.60

 matching

0.59

 arsen

0.59

 DEFENSE

0.58

 Reaction

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.