Idanwo ti o sopọ mọ itọsọna · Alabọde Ipele
ServiceNow ati StarCoder adanwo
Review the original StarCoderBase model, its training corpus, 8K context, fill-in-the-middle objective, and license conditions.
Ibeere 1 ti 8
Awọn ajo meji wo ni o ṣe akoso iṣẹ akanṣe BigCode ti o ṣe agbekalẹ StarCoder?
Awotẹlẹ gbogbo ibeere8 ibeere
Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.
- Awọn ajo meji wo ni o ṣe akoso iṣẹ akanṣe BigCode ti o ṣe agbekalẹ StarCoder?
- Iwadi ServiceNow ati Oju Famọra
- Google ati Meta
- OpenAI ati Microsoft
- Amazon ati IBM
- Which dataset was used for StarCoderBase pretraining?
- Pile naa
- Awọn akopọ
- Wọpọ Jijo
- LAION-5B
- Ilana ikẹkọ wo ni o jẹ ki StarCoder pari koodu nipa lilo ipo ni ẹgbẹ mejeeji ti kọsọ?
- Ẹkọ imudara lati awọn esi eniyan
- Awoṣe ede bojuboju
- Kun-ni-ni-Arin
- Ìfilọlẹ pq-ti-ero
- Ni aijọju bawo ni awoṣe flagship StarCoderBase ṣe tobi?
- 1,5 bilionu paramita
- 70 bilionu paramita
- 175 bilionu paramita
- 15,5 bilionu paramita
- Iru iṣapeye akiyesi wo ni StarCoder lo lati dinku iranti ati iyara iyara?
- Olona-ibeere akiyesi
- Sisun-window akiyesi
- Olofofo adalu amoye
- Ifojusi laini
- Labẹ iru iwe-aṣẹ wo ni a ti tu silẹ StarCoder lati ṣe iwuri fun lilo lodidi?
- GPL v3
- OpenRAIL-M lodidi AI iwe-ašẹ
- Iwe-aṣẹ iṣowo ohun-ini
- MIT iwe-ašẹ
- What context length did the original StarCoder paper report?
- 800 tokens
- 80,000 tokens
- 8,000 tokens
- One million tokens
- How was StarCoder produced from StarCoderBase?
- It was trained only on image captions
- It replaced The Stack with a speech corpus
- It was created by reducing the parameter count to 1.5B
- It was fine-tuned on 35 billion Python tokens