INDEX

Explanations

words related to connections and relationships

oai_token-act-pair · gpt-3.5-turbo

connections and relationships among people, concepts, or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

cheat

-0.73

umer

-0.69

grim

-0.68

anchez

-0.67

veland

-0.66

gery

-0.64

IGN

-0.64

gun

-0.63

ĳå£«

-0.63

odder

-0.62

POSITIVE LOGITS

 between

1.14

between

1.07

 Between

0.94

 sexes

0.89

 partners

0.87

 partner

0.85

 twins

0.83

 seamlessly

0.81

 disparate

0.77

SHIP

0.76

Activations Density 0.330%

words related to connections and relationships

connections and relationships among people, concepts, or entities

No Comments

No Known Activations