INDEX

Explanations

terms related to computing processes and components in software

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.hook_resid_post

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

Ĳ

-2.49

ľĵ

-2.43

ĺ

-2.35

Į

-2.34

¥

-2.20

Ļª

-2.15

´

-2.07

Ĩ

-2.05

ł

-2.04

°

-2.02

POSITIVE LOGITS

xe

1.83

 manually

1.79

 themselves

1.77

 typically

1.62

onas

1.57

 indefinitely

1.52

 simultaneously

1.46

 unless

1.46

 periodically

1.45

 Alternatively

1.45

Activations Density 7.669%

terms related to computing processes and components in software

No Comments

No Known Activations