INDEX

Explanations

the word "that" 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

ees

-1.86

ings

-1.68

 whom

-1.61

idades

-1.61

 quarters

-1.58

iously

-1.56

entes

-1.44

ieties

-1.40

liness

-1.40

eenth

-1.38

POSITIVE LOGITS

Ī

2.13

<|outofrange|>

2.12

<|outofrange|>

2.12

↵

2.12

<|outofrange|>

2.12

↵

2.12

↵↵↵

2.12

č↵

2.12

<|outofrange|>

2.12

Activations Density 0.233%

the word "that" in various contexts

No Comments

No Known Activations