{"links":[{"source":"/notes/bpe/","target":"/notes/token/"},{"source":"/notes/bpe/","target":"/notes/llm/"},{"source":"/notes/decoding/","target":"/notes/tokenizer/"},{"source":"/notes/decoding/","target":"/notes/token/"},{"source":"/notes/decoding/","target":"/notes/encoding/"},{"source":"/notes/encoding/","target":"/notes/tokenizer/"},{"source":"/notes/encoding/","target":"/notes/token/"},{"source":"/notes/vocabulary/","target":"/notes/token/"},{"source":"/notes/vocabulary/","target":"/notes/tokenizer/"},{"source":"/notes/rlhf/","target":"/notes/reinforcement-learning-from-human-feedback/"},{"source":"/notes/distillation/","target":"/notes/large-language-model/"},{"source":"/notes/distillation/","target":"/notes/supervised-fine-tuning/"},{"source":"/notes/distillation/","target":"/notes/llm/"},{"source":"/notes/reinforcement-learning-from-human-feedback/","target":"/notes/rlhf/"},{"source":"/notes/reinforcement-learning-from-human-feedback/","target":"/notes/reinforcement-learning/"},{"source":"/notes/reinforcement-learning-from-human-feedback/","target":"/notes/reasoning/"},{"source":"/notes/inference-compute-scaling/","target":"/notes/inference-time-compute-scaling/"},{"source":"/notes/inference-time-compute-scaling/","target":"/notes/inference-compute-scaling/"},{"source":"/notes/inference-time-compute-scaling/","target":"/notes/test-time-scaling/"},{"source":"/notes/inference-time-compute-scaling/","target":"/notes/large-language-model/"},{"source":"/notes/inference-time-compute-scaling/","target":"/notes/reasoning/"},{"source":"/notes/reasoning/","target":"/notes/large-language-model/"},{"source":"/notes/reasoning/","target":"/notes/llm-training-pipeline/"},{"source":"/notes/reasoning/","target":"/notes/inference-time-compute-scaling/"},{"source":"/notes/reasoning/","target":"/notes/inference-compute-scaling/"},{"source":"/notes/reasoning/","target":"/notes/test-time-scaling/"},{"source":"/notes/reasoning/","target":"/notes/reinforcement-learning/"},{"source":"/notes/reasoning/","target":"/notes/rl/"},{"source":"/notes/reasoning/","target":"/notes/distillation/"},{"source":"/notes/reinforcement-learning/","target":"/notes/rl/"},{"source":"/notes/reinforcement-learning/","target":"/notes/large-language-model/"},{"source":"/notes/reinforcement-learning/","target":"/notes/reasoning/"},{"source":"/notes/reinforcement-learning/","target":"/notes/inference-time-compute-scaling/"},{"source":"/notes/reinforcement-learning/","target":"/notes/reinforcement-learning-from-human-feedback/"},{"source":"/notes/reinforcement-learning/","target":"/notes/rlhf/"},{"source":"/notes/reinforcement-learning/","target":"/notes/preference-tuning/"},{"source":"/notes/rl/","target":"/notes/reinforcement-learning/"},{"source":"/notes/test-time-scaling/","target":"/notes/inference-time-compute-scaling/"},{"source":"/notes/token/","target":"/notes/tokenizer/"},{"source":"/notes/token/","target":"/notes/bpe/"},{"source":"/notes/tokenizer/","target":"/notes/llm/"},{"source":"/notes/tokenizer/","target":"/notes/token/"},{"source":"/notes/tokenizer/","target":"/notes/encoding/"},{"source":"/notes/tokenizer/","target":"/notes/decoding/"},{"source":"/notes/llm-training-pipeline/","target":"/notes/llm/"},{"source":"/notes/llm-training-pipeline/","target":"/notes/pre-training/"},{"source":"/notes/llm-training-pipeline/","target":"/notes/post-training/"},{"source":"/notes/preference-tuning/","target":"/notes/reinforcement-learning-from-human-feedback/"},{"source":"/notes/preference-tuning/","target":"/notes/rlhf/"},{"source":"/notes/preference-tuning/","target":"/notes/supervised-fine-tuning/"},{"source":"/notes/supervised-fine-tuning/","target":"/notes/large-language-model/"},{"source":"/notes/supervised-fine-tuning/","target":"/notes/preference-tuning/"},{"source":"/notes/chain-of-thought-cot/","target":"/notes/llm/"},{"source":"/notes/chain-of-thought-cot/","target":"/notes/reasoning/"},{"source":"/notes/agentic-loop/","target":"/notes/ai-agent/"},{"source":"/notes/ai-agent/","target":"/notes/large-language-model/"},{"source":"/notes/ai-agent/","target":"/notes/agentic-loop/"},{"source":"/notes/fine-tuning/","target":"/notes/large-language-model/"},{"source":"/notes/fine-tuning/","target":"/notes/supervised-fine-tuning/"},{"source":"/notes/fine-tuning/","target":"/notes/preference-tuning/"},{"source":"/notes/hallucination/","target":"/notes/large-language-model/"},{"source":"/notes/large-language-model/","target":"/notes/token/"},{"source":"/notes/large-language-model/","target":"/notes/parameters/"},{"source":"/notes/large-language-model/","target":"/notes/llm-training-pipeline/"},{"source":"/notes/large-language-model/","target":"/notes/hallucination/"},{"source":"/notes/parameters/","target":"/notes/large-language-model/"},{"source":"/notes/post-training/","target":"/notes/pre-training/"},{"source":"/notes/pre-training/","target":"/notes/large-language-model/"},{"source":"/notes/pre-training/","target":"/notes/token/"},{"source":"/notes/llm/","target":"/notes/large-language-model/"}],"nodes":[{"id":"/notes/bpe/","section":"notes","tags":["AI","machinelearning","LLM","nlp","tokenization"],"title":"BPE"},{"id":"/notes/decoding/","section":"notes","tags":["AI","machinelearning","LLM","nlp"],"title":"decoding"},{"id":"/notes/encoding/","section":"notes","tags":["AI","machinelearning","LLM","nlp"],"title":"encoding"},{"id":"/notes/vocabulary/","section":"notes","tags":["AI","machinelearning","LLM","nlp"],"title":"vocabulary"},{"id":"/notes/rlhf/","section":"notes","tags":["AI","LLM","reinforcementlearning"],"title":"RLHF"},{"id":"/notes/distillation/","section":"notes","tags":["AI","LLM","reasoning"],"title":"distillation"},{"id":"/notes/reinforcement-learning-from-human-feedback/","section":"notes","tags":["AI","LLM","machinelearning","reinforcementlearning"],"title":"reinforcement learning from human feedback"},{"id":"/notes/inference-compute-scaling/","section":"notes","tags":["AI","LLM"],"title":"inference-compute scaling"},{"id":"/notes/inference-time-compute-scaling/","section":"notes","tags":["AI","LLM"],"title":"inference-time compute scaling"},{"id":"/notes/reasoning/","section":"notes","tags":["AI","LLM"],"title":"reasoning"},{"id":"/notes/reinforcement-learning/","section":"notes","tags":["AI","LLM","reinforcementlearning"],"title":"reinforcement learning"},{"id":"/notes/rl/","section":"notes","tags":["AI","LLM","reinforcementlearning"],"title":"RL"},{"id":"/notes/test-time-scaling/","section":"notes","tags":["AI","LLM"],"title":"test-time scaling"},{"id":"/notes/token/","section":"notes","tags":["AI","machinelearning","LLM","nlp"],"title":"token"},{"id":"/notes/tokenizer/","section":"notes","tags":["AI","machinelearning","LLM","nlp"],"title":"tokenizer"},{"id":"/notes/llm-training-pipeline/","section":"notes","tags":["LLM","AI","deeplearning"],"title":"LLM training pipeline"},{"id":"/notes/preference-tuning/","section":"notes","tags":["AI","machinelearning","LLM","deeplearning"],"title":"preference tuning"},{"id":"/notes/supervised-fine-tuning/","section":"notes","tags":["AI","machinelearning","LLM","deeplearning"],"title":"supervised fine-tuning"},{"id":"/notes/chain-of-thought-cot/","section":"notes","tags":["LLM","AI","agents"],"title":"chain-of-thought"},{"id":"/notes/agentic-loop/","section":"notes","tags":["AI","agents","computerscience","automation","architecture","machinelearning"],"title":"agentic loop"},{"id":"/notes/ai-agent/","section":"notes","tags":["AI","agents","computerscience","automation","architecture","LLM"],"title":"AI agent"},{"id":"/notes/fine-tuning/","section":"notes","tags":["AI","machinelearning","LLM","deeplearning"],"title":"fine-tuning"},{"id":"/notes/hallucination/","section":"notes","tags":["LLM","AI"],"title":"hallucination"},{"id":"/notes/large-language-model/","section":"notes","tags":["AI","machinelearning","nlp","deeplearning","LLM"],"title":"Large Language Model"},{"id":"/notes/parameters/","section":"notes","tags":["LLM","AI","deeplearning","machinelearning"],"title":"parameters"},{"id":"/notes/post-training/","section":"notes","tags":["AI","machinelearning","LLM","deeplearning"],"title":"post-training"},{"id":"/notes/pre-training/","section":"notes","tags":["AI","machinelearning","LLM","deeplearning"],"title":"pre-training"},{"id":"/posts/hello-world/","section":"posts","tags":["helloworld"],"title":"Hello World"},{"id":"/notes/llm/","section":"notes","tags":["LLM","AI","glossary"],"title":"LLM"}]}