Archive8 [λ²μ] The History of Open-Source LLMs: Part β ‘. Better Base Models The History of Open-Source LLMs: Part β . Early days The History of Open-Source LLMs: Part β ‘. Better Base Models The History of Open-Source LLMs: Part β ’. Imitations and alignment β κΈμ μ½μ λ λͺ¨λ μ΄λ―Έμ§μ μΆμ²λ μλ¬Έμ λλ€. μ΄κΈ° μ€νμμ€ LLM - 곡κ°λμ§ μμ μ¬μ νμ΅ λͺ¨λΈλ€κ³Ό λΉκ΅νλ©΄ μ±λ₯μ΄ λ§μ΄ λ¨μ΄μ§λ€λ λ¨μ μ΄ μ‘΄μ¬ν¨ 1) LLM νμ΅ νμ΄νλΌμΈ β λλμ μμ λ°μ΄ν°λ₯Ό μ΄μ©ν΄ λͺ¨λΈ μ¬μ νμ΅ β‘ SFTμ RLHF κ°μ κΈ°μ μ μ΄μ©ν΄ alignment μν β’ LLMμ νΉμ νμ€ν¬μ μ μ©νκΈ° μν΄ νμΈ νλ λλ in-context learning μν.. 2023. 10. 25. [λ²μ] The History of Open-Source LLMs: Part β . Early days μ°κ΅¬μ€ μ€λΉ κ° κ³΅μ ν΄μ€ μ€νμμ€ LLM λ°μ κ³Όμ μ λν κΈμ μ½κ³ λ²μ λ° μΆκ° 곡λΆν λ΄μ©μ μ 리ν΄λ³Έλ€. The History of Open-Source LLMs: Part β . Early days The History of Open-Source LLMs: Part β ‘. Better Base Models The History of Open-Source LLMs: Part β ’. Imitations and alignment β κΈμ μ½μ λ λͺ¨λ μ΄λ―Έμ§μ μΆμ²λ μλ¬Έμ λλ€. LLM λ±μ₯ λ°°κ²½ - μΈμ΄ λͺ¨λΈ μ체λ μμ¬κ° μ€λ λμ§λ§ self-supervised pre-trainingκ³Ό in-context learningμ μ‘°ν©νμ¬ μ¬λ¬ νμ€ν¬μμ μΈμ κΉμ few-shot learning μ±λ₯μ λ³΄μΈ GP.. 2023. 10. 23. [λ²μ] Micro, Macro & Weighted Averages of F1 Score, Clearly Explained π¬ μ΅λν 맀λλ½κ² ν΄μνκ³ μ λ Έλ ₯νμ§λ§ μ΄μν λ¬Έμ₯μ΄ μμ μ μμ΅λλ€. νΌλλ°±μ μΈμ λ νμμ λλ€ π μλ³Έ κΈ μ£Όμ: https://towardsdatascience.com/micro-macro-weighted-averages-of-f1-score-clearly-explained-b603420b292f Micro, Macro & Weighted Averages of F1 Score, Clearly Explained Understanding the concepts behind the micro average, macro average and weighted average of F1 score in multi-class classification towardsdatascience.com F1 Score(.. 2022. 12. 20. [λ²μ] Foundations of NLP Explained Visually: Beam Search, How It Works π¬ μ΅λν 맀λλ½κ² ν΄μνκ³ μ λ Έλ ₯νμ§λ§ μ΄μν λ¬Έμ₯μ΄ μμ μ μμ΅λλ€. νΌλλ°±μ μΈμ λ νμμ λλ€ π μλ³Έ κΈ μ£Όμ : https://towardsdatascience.com/foundations-of-nlp-explained-visually-beam-search-how-it-works-1586b9849a24 Foundations of NLP Explained Visually: Beam Search, How it Works A Gentle Guide to how Beam Search enhances predictions, in Plain English towardsdatascience.com NLP λͺ¨λΈμ Output μμ± λ°©λ² κΈ°κ³ λ²μ μμ μ μμ£Ό μ¬μ©λλ sequence-to-sequence λͺ¨.. 2022. 8. 1. [λ²μ] Word2Vec Research Paper Explained π¬ μ΅λν 맀λλ½κ² ν΄μνκ³ μ λ Έλ ₯νμ§λ§ μ΄μν λ¬Έμ₯μ΄ μμ μ μμ΅λλ€. νΌλλ°±μ μΈμ λ νμμ λλ€ π μλ³Έ κΈ μ£Όμ : https://towardsdatascience.com/word2vec-research-paper-explained-205cb7eecc30 Word2Vec Research Paper Explained An Intuitive understanding and explanation of the word2vec model. towardsdatascience.com μλ‘ λ§μ NLP νμ© μ¬λ‘μμ, λ¨μ΄λ μλ‘ κ°μ κ΄κ³λ₯Ό ν¬μ°©νμ§ μλ μ-ν« μΈμ½λ©μΌλ‘ ννλλ€. μ-ν« μΈμ½λ©μ μ¬μ©νλ μ΄μ λ "κ°λ¨ν¨, κ²¬κ³ ν¨κ³Ό λλμ λ°μ΄ν°λ‘ νλ ¨λ κ°λ¨ν λͺ¨λΈμ΄ μλμ λ°μ΄ν°λ‘ νλ ¨λ 볡μ‘ν λͺ¨λΈλ³΄λ€.. 2022. 7. 4. [λ²μ] Introduction to Stemming and Lemmatization π¬ μ΅λν 맀λλ½κ² ν΄μνκ³ μ λ Έλ ₯νμ§λ§ μ΄μν λ¬Έμ₯μ΄ μμ μ μμ΅λλ€. νΌλλ°±μ μΈμ λ νμμ λλ€ π μλ³Έ κΈ μ£Όμ : https://medium.com/geekculture/introduction-to-stemming-and-lemmatization-nlp-3b7617d84e65 μμ°μ΄μ²λ¦¬(Natural Language Processing, NLP) ν μ€νΈ λ°μ΄ν°λ λ€μν μμ€λ‘λΆν° μ»μ μ μλ€. μ°λ¦¬μ λͺ©νλ νμ€ν¬μ κ΄λ ¨μλ μμ€ κ³ μ μ λ§ν¬μ μ΄λ κ΅¬λ¬Έμ΄ μλ νλ¬Έμ μΆμΆνλ κ²μ΄λ€. ꡬλμ , λλ¬Έμμ κ°μ μΈμ΄μ λͺλͺ νΉμ±κ³Ό "a"/"of"/"the"μ κ°μ μΌλ°μ μΈ λ¨μ΄λ€μ΄ λ¬Έμμ ꡬ쑰 μ 곡μ λμμ μ£ΌκΈ΄ νμ§λ§ λ§μ μλ―Έλ₯Ό μ£Όμ§λ μλλ€. λ°λΌμ ν μ€νΈ λ°μ΄ν°λ₯Ό λΆμνμ¬ μμ°μ΄ μ²λ¦¬ νμ΄.. 2022. 4. 1. μ΄μ 1 2 λ€μ