INDEX

Explanations

phrases related to law enforcement and public safety

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

pai

-0.67

rome

-0.66

é¾įå

-0.66

apore

-0.65

å¥³

-0.65

haar

-0.65

TG

-0.64

ually

-0.63

ega

-0.63

"{

-0.63

POSITIVE LOGITS

 extingu

0.94

 destro

0.85

 exting

0.85

 fury

0.83

 extinguished

0.80

 inciner

0.79

 grill

0.78

 fires

0.76

 brim

0.76

 cannons

0.73

Activations Density 0.048%

phrases related to law enforcement and public safety

No Comments

No Known Activations