INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

pei

-0.76

fol

-0.72

 Amen

-0.70

isec

-0.68

 Scher

-0.67

imens

-0.67

aido

-0.67

 perfume

-0.64

allo

-0.63

okin

-0.62

POSITIVE LOGITS

IRT

0.74

arth

0.64

ashington

0.63

outheast

0.62

abolic

0.62

Ħ¢

0.60

ERT

0.60

 minimized

0.59

abetic

0.58

 Intellectual

0.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.