INDEX

Explanations

instances of the word "whose."

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

thems

-1.49

mine

-1.47

Category

-1.43

:`

-1.38

represented

-1.36

shire

-1.35

:]

-1.34

 considering

-1.34

gov

-1.33

CMS

-1.31

POSITIVE LOGITS

Į

3.02

ª

2.90

Ļ

2.82

İ

2.81

³

2.70

·¸

2.69

¨

2.69

¥

2.68

¸

2.67

¶

2.66

Activations Density 0.223%

instances of the word "whose."

No Comments

No Known Activations