A continuaciónSiguiente guía
Leyes de escala para redes neuronales
Fundamentos
GUÍA de IA en idiomas
Chinchilla studied how to allocate a training-compute budget between parameters and training tokens.
Escuche esta guíaNarrado por IA · ~2 min
Its fitted results favor scaling both together in the evaluated regime; that objective is different from minimizing total deployment cost or guaranteeing one token ratio for every dataset.
The 2022 study fitted loss behavior across model and data budgets and estimated training-compute-optimal parameter and token allocations. The finding that both quantities should grow in similar proportions means comparable scaling exponents, not equal numerical counts. Chinchilla used about seventy billion parameters and 1.4 trillion training tokens, giving a twenty-token-per-parameter example. It was compared with the much larger Gopher under a comparable training-compute budget. The study argued that many evaluated contemporary large models were undertrained relative to the estimated optimum. These relationships are fitted results that depend on the data and setup; a fixed ratio is not a universal design law. LLaMA's paper discusses training smaller models longer as a way to consider inference efficiency, a different objective from training compute alone.
Comparing the 70-billion-parameter Chinchilla experiment with the 280-billion-parameter Gopher baseline.
Using fitted training-budget relationships as a starting point rather than a universal ratio rule.
Distinguishing a from-scratch training optimum from a smaller model chosen for repeated inference.
Evaluating whether parameters or data provide more value under a defined compute and quality objective.
Scaling research continues to examine data quality, repeated data, model architecture, and the cost of training versus serving. State the objective and evaluation conditions before calling an allocation optimal. A useful empirical trend is different from a guarantee that one parameter-token recipe dominates every workload.
Boletín gratuito
Tres noticias verificadas sobre IA cada mañana de lunes a viernes, explicadas con claridad. Gratis para siempre y sin anuncios.
Un correo cada día laborable. Date de baja con un clic. Nunca vendemos ni compartimos tu dirección.
Ponte a prueba
Corrección inmediata en cada respuesta y un certificado para compartir, con un ID verificable, cuando apruebas un curso.
Apoya la educación gratuita sobre IA. AI Understanding es una organización sin fines de lucro 501(c)(3): sin anuncios ni muros de pago, nunca. Hacer una donación
Chinchilla studied how to allocate a training-compute budget between parameters and training tokens. Its fitted results favor scaling both together in the evaluated regime; that objective is different from minimizing total deployment cost or guaranteeing one token ratio for every dataset.
Similar proportional growth is not a literal one-token-per-parameter rule.
1.4 trillion divided by seventy billion equals twenty; this example is not a universal optimum for every setup.
El Chinchilla de 70B de parámetros venció al Gopher de 280B de parámetros de DeepMind usando el mismo cálculo, porque se entrenó con muchos más datos.
Muchos modelos grandes de esa época no estaban suficientemente entrenados, lo que significa que habrían funcionado mejor con muchos más datos para el recuento de parámetros.
El cálculo de entrenamiento (FLOP) es aproximadamente proporcional al número de parámetros multiplicado por el número de tokens de entrenamiento.
sigue aprendiendo
Más guías seleccionadas para este tema.
A continuaciónSiguiente guía
Leyes de escala para redes neuronales
Fundamentos