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

 fumes

-0.72

edia

-0.72

 fashioned

-0.67

 masc

-0.67

 blockers

-0.66

zoo

-0.65

agements

-0.64

enders

-0.64

icularly

-0.64

wed

-0.63

POSITIVE LOGITS

yan

0.81

ton

0.77

orian

0.74

roid

0.73

sson

0.72

yang

0.72

ĵ

0.71

ennial

0.70

nea

0.69

£ı

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.