#AAT3222IGV-2.8-T1

[icbros]

Meta launched the AI language model LLaMA, a large language model with 65 billion parameters

Meta announced on Friday local time that it will launch a new large-scale language model based on artificial intelligence (AI) for the research community, and join the AI competition with Microsoft, Google and other companies stimulated by ChatGPT. Meta's LLaMA is the abbreviation of "Large Language Model Meta AI". It can be provided to government, community and academic researchers and entity workers under non-commercial license. The company will provide the underlying code for users to use, so users can adjust the model and use it for use cases related to research. Meta said that the model requires "much less computational power". It is reported that the company is developing LLaMA with various parameters (7B, 13B, 33B and 65B). Among them, LLaMA 65B and LLaMA 33B have been trained on 1.4 trillion tokens, while the smallest model LLaMA 7B has also been trained on 1 trillion tokens. Like other large language models, LLaMA works by taking a series of words as "input" and predicting the next word to recursively generate text. For this model, Meta selects texts from 20 languages with the largest number of users for training, focusing on Latin and Cyrillic alphabet. Of course, like other models, LLaMA also faces the challenges of prejudice, toxic comments and hallucinations. Meta still needs to do more research to solve the shortcomings of such language models. Meta said that LLaMA, as a basic model, is designed to be multifunctional and can be applied to many different use cases, rather than a fine-tuning model designed for specific tasks. Through the open source LLaMA code, other researchers can more easily find new ways to limit or eliminate these problems. Meta also provides a set of benchmark evaluation criteria for evaluating model bias and toxicity in this paper to show the limitations of the model and support researchers' further research in this key area. It is worth mentioning that Meta also launched the large-scale language model OPT-175B in May last year. The project is also aimed at researchers, which forms the basis of the new iteration of its chat robot blenterbot. Later, the company also launched a model called "Galactica", which was said to be able to write scientific articles and solve mathematical problems, but its demo version was later removed from the shelf because it repeatedly generated "sounds authoritative" content.

ICBROS TECHNOLOGY LIMITED is a worldwide professional electronic components distributor.

Our hot chip models today:

AAT7412-S5-T SOT23-5 AAT2138IWO-0.6-T1 TDFN33-14 AAT1118-T1-T TSSOP16 AAT7359B-Q19-T WSON AAT1013A DIP8 AAT3222IGV-1.8-T1 SOT23-5 AAT3222IGV-2.8-T1 SOT23-5 AAT3222IGV-3.0-T1 SOT23-5 316B102 DIP16 316A471 CDIP16 MC80F0104D SOP MC80F0504M SOP16 5SDA11D1702 MODULE SHHS001A3B641 MOUDLE QBVW025A0B41-HZ MOUDLE 2611 SOP8 S3N250R2003PFF AXH005A0X-SR SMD JNW350R641Z 18 CM2520101J3B SMD TEST001 DIP SR0604151KLB SMD YF840 TO-220 LPF3015T-4R7M SMD LPF7045T-3R3M-C SMD LMC2012TP-270J SMD LMC2012TP-331J SMD LMC2012TP-221J SMD LMC2012TP-331G SMD LPF2010T-100M SMD LMC1608TP-R11G SMD LMC1608TP-39NJ SMD LMC1608TP-2N2J SMD LMC1608TP-27NJ SMD LMC1608TP-33NJ SMD LMC1608TP-56NJ SMD AL03TB101K DO-15 LMC1608TP-R15J SMD LPF1280T-220M SMD LPF3010T-220M SMD LMC2012TP-391G SMD LPF6524T-1R0M SMD LPF1280T-222M SMD LPF8044T-4R7M SMD LPF3015T-1R5N-C SMD LPF2810T-221M SMD LPS252010T-100M SMD LPF2010T-4R7M SMD LMC1608TP-R27G SMD LMC1005TP-27NJ SMD LMC1608TP-R33J SMD LMC2012TP-391J SMD0805 LPF2015-1R0N SMD LPF2015T-6R8M SMD LMF1005TP-101K SMD LMC1608TP-15NJ SMD LPF5030T-4R7M SMD LPF1280T-100M-1 SMD LMC2012TP-271J SMD LMC2012TP-471G SMD LPF3513T-100M SMD LPH4018T-2R2M SMD LPF4017T-4R7N SMD LMC1608TP-R10J SMD LPF1280 SMD A32M001 QFN AB801-01 BGA AS44CE375ACEF SOP16 AS44CE374T1AEFR2 SOP16 ASC373CAEF SOP16 AS44CE375ACEFR2 SOIC16 AS44CC375AVEF SOP AS44CC374T1AEF SOP AS44CC373CAEF SOP16 AS44CC373CASEF SOP16 S-8225AAH-TCT1U TSSOP16 S-812C33BMC-C4NT2U SOT23-5 S-8252ABK-I6T1U SNT-6A S-8261DAX-M6T1U SOT23-6 S-19110EAPH-M6T1U4 SOT23-6 S-8520D50MC-BWJT2G 18 MC93F5632LBN QFP44 MC80F0316GP DIP28 ADAM43P1108G SOP8 MC97F1204SMBN SOP16 MC96F8316D SOP MC93CV402U QFN MC96F1206USBN QFN20 MC80C0316-MC019 SOP D009S SOP 80F0504D SOP16 MC80F0504D SOP16 ADAM22P20P SOP20 MC97F1204SRBN TSSOP-20 HMS87C1716BKP DIP42 GMS81524BT DIP64 MC80F7708QP QFP44 80F0504M SOP16

Email：jammy@icbros.com HK Tel：+852 44058386 Whatsapp：+852 44058386 Skype：jammy@icbros.com Web:   www.icbros.com