INDEX

Explanations

names of professions or roles

oai_token-act-pair · gpt-3.5-turbo

mentions of positive attributes or characteristics related to people, teams, and performance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Downloadha

-0.69

soDeliveryDate

-0.66

 dreaded

-0.65

 mandated

-0.64

 omin

-0.63

 relinqu

-0.63

owered

-0.63

 guiActiveUn

-0.63

 controvers

-0.63

NetMessage

-0.62

POSITIVE LOGITS

smanship

0.93

 luck

0.88

manship

0.87

 mate

0.83

 payoff

0.82

guy

0.81

 experience

0.78

bye

0.78

 performance

0.76

 achie

0.75

Activations Density 0.349%

names of professions or roles

mentions of positive attributes or characteristics related to people, teams, and performance

No Comments

No Known Activations