SeterusnyaPanduan seterusnya
Undang-undang Penskalaan untuk Rangkaian Neural
Asas
PANDUAN AI Bahasa
Chinchilla studied how to allocate a training-compute budget between parameters and training tokens.
Dengar panduan iniAI-diceritakan · ~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.
Newsletter percuma
Tiga cerita AI yang disahkan setiap pagi hari bekerja, ditulis dalam Bahasa Inggeris biasa. Percuma selamanya, tiada iklan.
Satu e-mel setiap hari bekerja. Berhenti melanggan dalam satu klik. Kami tidak pernah menjual atau berkongsi alamat anda.
Uji diri anda
Maklum balas segera pada setiap jawapan, dan sijil yang boleh dikongsi dengan ID yang boleh disahkan sebaik sahaja anda lulus kursus.
Sokong pendidikan AI percuma. AI Understanding ialah pertubuhan bukan keuntungan 501(c)(3) — tiada iklan, tiada bayaran dinding, selamanya. Buat derma
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.
Chinchilla 70B parameter mengalahkan Gopher 280B parameter DeepMind sendiri menggunakan pengiraan yang sama, kerana ia melatih lebih banyak data.
Banyak model besar pada era itu kurang terlatih, bermakna mereka akan menunjukkan prestasi yang lebih baik dengan lebih banyak data untuk kiraan parameter mereka.
Pengiraan latihan (FLOP) adalah lebih kurang berkadar dengan bilangan parameter didarab dengan bilangan token latihan.
Teruskan belajar
Lebih banyak panduan dipilih untuk topik ini
SeterusnyaPanduan seterusnya
Undang-undang Penskalaan untuk Rangkaian Neural
Asas