INDEX

Explanations

action words or intentions related to improvement and change

oai_token-act-pair · gpt-3.5-turbo

phrases expressing intentions or desires to improve or change situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

spection

-0.69

raped

-0.68

fax

-0.67

ominated

-0.67

ELD

-0.66

 Stern

-0.64

 Trouble

-0.63

urbed

-0.63

rieved

-0.63

ears

-0.63

POSITIVE LOGITS

 maximize

1.46

 minimize

1.40

 simplify

1.37

 create

1.35

 educate

1.34

 ensure

1.33

 prioritize

1.30

 capitalize

1.29

 empower

1.28

 optimize

1.28

Activations Density 0.405%

action words or intentions related to improvement and change

phrases expressing intentions or desires to improve or change situations

No Comments

No Known Activations