INDEX

Explanations

instances of the word "to" in various contexts

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hole

-1.55

 onwards

-1.49

 erupted

-1.42

CA

-1.41

patrick

-1.39

 onward

-1.38

arias

-1.37

iorari

-1.37

ry

-1.36

icular

-1.36

POSITIVE LOGITS

↵

2.47

↵

2.47

↵

2.47

↵↵

2.47

č↵

2.47

↵

2.47

<|outofrange|>

2.47

↵

2.47

↵↵↵

2.47

↵↵

2.47

Activations Density 0.661%

instances of the word "to" in various contexts

No Comments

No Known Activations