INDEX

Explanations

repetitions of the word "more."

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

 itself

-1.89

rin

-1.63

ORY

-1.57

rans

-1.51

abolism

-1.46

eur

-1.46

ilization

-1.45

ray

-1.38

agement

-1.37

 center

-1.36

POSITIVE LOGITS

»¿

2.35

↵

2.05

č↵

2.05

↵↵

2.05

↵↵↵

2.05

↵

2.05

↵↵

2.05

<|outofrange|>

2.05

<|outofrange|>

2.05

Activations Density 0.234%

repetitions of the word "more."

No Comments

No Known Activations