INDEX

Explanations

phrases related to discussing various topics in interviews or articles

oai_token-act-pair · gpt-3.5-turbo

phrases focusing on discussion about various topics and their details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

strom

-0.79

geon

-0.73

IX

-0.72

stretched

-0.70

wei

-0.70

cour

-0.69

ranged

-0.69

iva

-0.68

ario

-0.68

elaide

-0.67

POSITIVE LOGITS

 importance

1.39

 dangers

1.37

 significance

1.28

 pitfalls

1.28

 origins

1.24

 virtues

1.23

 difficulties

1.20

 evils

1.19

 genesis

1.18

 merits

1.16

Activations Density 0.283%

phrases related to discussing various topics in interviews or articles

phrases focusing on discussion about various topics and their details

No Comments

No Known Activations