https://www.fanfani.net/tutorials/inside-the-machines/2026-06-30T22:18:15+02:00https://www.fanfani.net/tutorials/diffusion-lessons/lesson-01-the-big-picture/https://www.fanfani.net/tutorials/llm-lessons/lesson-01-the-big-picture/https://www.fanfani.net/tutorials/moe-lessons/lesson-01-the-big-picture/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-01-the-big-picture/https://www.fanfani.net/tutorials/inside-the-machines/chapter-01-what-these-machines-are/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-01-why-two-stacks/https://www.fanfani.net/tutorials/llm-lessons/2026-06-30T10:52:57+02:00https://www.fanfani.net/tutorials/moe-lessons/lesson-02-from-ffn-to-experts/https://www.fanfani.net/tutorials/inside-the-machines/chapter-02-from-words-to-numbers/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-02-the-encoder-stack/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-02-forward-process/https://www.fanfani.net/tutorials/llm-lessons/lesson-02-tokenization-and-embeddings/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-02-tokenization-and-embeddings/https://www.fanfani.net/tutorials/diffusion-lessons/2026-06-30T10:52:57+02:00https://www.fanfani.net/tutorials/inside-the-machines/chapter-03-attention/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-03-bidirectional-self-attention/https://www.fanfani.net/tutorials/llm-lessons/lesson-03-self-attention/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-03-the-decoder-stack/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-03-reverse-process/https://www.fanfani.net/tutorials/moe-lessons/lesson-03-the-router-and-gating/https://www.fanfani.net/tutorials/encoder-deep-dive/2026-06-30T11:57:28+02:00https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-04-cross-attention/https://www.fanfani.net/tutorials/moe-lessons/lesson-04-sparse-computation-and-combine/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-04-the-encoder-block-and-stack/https://www.fanfani.net/tutorials/inside-the-machines/chapter-04-decoder-models/https://www.fanfani.net/tutorials/llm-lessons/lesson-04-the-transformer-block/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-04-variational-lower-bound/https://www.fanfani.net/tutorials/moe-lessons/2026-06-30T13:40:14+02:00https://www.fanfani.net/tutorials/moe-lessons/lesson-05-load-balancing-and-auxiliary-losses/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-05-masked-language-modeling/https://www.fanfani.net/tutorials/llm-lessons/lesson-05-output-head-and-objective/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-05-the-encoder-decoder-block/https://www.fanfani.net/tutorials/inside-the-machines/chapter-05-encoder-models/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-05-simplified-objective/https://www.fanfani.net/tutorials/encoder-decoder-lessons/2026-06-30T15:09:43+02:00https://www.fanfani.net/tutorials/moe-lessons/lesson-06-capacity-and-token-dropping/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-06-representations-to-predictions/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-06-prefix-lm/https://www.fanfani.net/tutorials/llm-lessons/lesson-06-the-loss-function/https://www.fanfani.net/tutorials/inside-the-machines/chapter-06-encoder-decoder-models/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-06-unet-architecture/https://www.fanfani.net/tutorials/llm-lessons/lesson-07-backpropagation-and-gradient-descent/https://www.fanfani.net/tutorials/moe-lessons/lesson-07-expert-parallelism/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-07-loss-and-evaluation/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-07-pretraining-objectives/https://www.fanfani.net/tutorials/inside-the-machines/chapter-07-mixture-of-experts/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-07-training-algorithm/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-08-ddpm-sampling/https://www.fanfani.net/tutorials/llm-lessons/lesson-08-optimizers-learning-rate-batches/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-08-pretraining-dynamics/https://www.fanfani.net/tutorials/inside-the-machines/chapter-08-diffusion-models/https://www.fanfani.net/tutorials/moe-lessons/lesson-08-training-dynamics-and-scaling/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-08-training-teacher-forcing/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-09-ddim-and-score/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-09-finetuning-downstream/https://www.fanfani.net/tutorials/moe-lessons/lesson-09-finetuning-moe/https://www.fanfani.net/tutorials/llm-lessons/lesson-09-finetuning-sft-rlhf/https://www.fanfani.net/tutorials/inside-the-machines/chapter-09-how-a-model-learns/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-09-inference/https://www.fanfani.net/tutorials/diffusion-lessons/lesson-10-conditioning-and-guidance/https://www.fanfani.net/tutorials/encoder-decoder-lessons/lesson-10-finetuning-and-distillation/https://www.fanfani.net/tutorials/llm-lessons/lesson-10-inference-in-depth/https://www.fanfani.net/tutorials/moe-lessons/lesson-10-inference-in-depth/https://www.fanfani.net/tutorials/encoder-deep-dive/lesson-10-practice-and-landscape/https://www.fanfani.net/tutorials/inside-the-machines/chapter-10-raising-a-helpful-assistant/https://www.fanfani.net/tutorials/encoder-deep-dive/appendix-a-encoder-family/https://www.fanfani.net/tutorials/diffusion-lessons/appendix-a-acronyms/https://www.fanfani.net/tutorials/llm-lessons/appendix-a-acronyms/https://www.fanfani.net/tutorials/encoder-decoder-lessons/appendix-a-formulas/https://www.fanfani.net/tutorials/inside-the-machines/chapter-11-how-models-run/https://www.fanfani.net/tutorials/llm-lessons/appendix-a-chinchilla-scaling-laws/https://www.fanfani.net/tutorials/encoder-decoder-lessons/appendix-b-architecture-comparison/https://www.fanfani.net/tutorials/encoder-deep-dive/appendix-b-encoder-vs-decoder/https://www.fanfani.net/tutorials/diffusion-lessons/appendix-b-formulas/https://www.fanfani.net/tutorials/inside-the-machines/chapter-12-the-whole-landscape/https://www.fanfani.net/tutorials/inside-the-machines/appendix-a-field-guide-to-models/https://www.fanfani.net/tutorials/llm-lessons/appendix-b-formulas/https://www.fanfani.net/tutorials/encoder-deep-dive/appendix-c-acronyms/https://www.fanfani.net/tutorials/encoder-decoder-lessons/appendix-c-acronyms/https://www.fanfani.net/tutorials/llm-lessons/appendix-c-attention-variants/https://www.fanfani.net/tutorials/inside-the-machines/appendix-b-comparison-cheat-sheets/https://www.fanfani.net/tutorials/inside-the-machines/appendix-c-attention-variants/https://www.fanfani.net/tutorials/llm-lessons/appendix-d-positional-encodings/https://www.fanfani.net/tutorials/inside-the-machines/appendix-d-positional-encodings/https://www.fanfani.net/tutorials/inside-the-machines/appendix-e-chinchilla-scaling/https://www.fanfani.net/tutorials/inside-the-machines/glossary/https://www.fanfani.net/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/ablation/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/adamw/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/agents/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/arithmetic-intensity/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/attention/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/batch-size/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/cot-faithfulness/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/context-engineering/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/continuous-batching/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/cross-entropy/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/data-mixtures/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/decoding/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/diffusion-lms/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/distillation/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/dpo/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/embeddings/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/emergent-abilities/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/evaluation/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/feed-forward-layers/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/flashattention/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/gradient-checkpointing/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/gradient-descent/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/grouped-query-attention/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/hallucination/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/induction-heads/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/inference-quantization/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/initialization/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/kl-divergence/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/kv-cache/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/ai-feedback/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/learning-rate-schedules/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/linear-attention/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/logits/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/long-context/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/lora/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/loss-spikes/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/memorization/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/mixed-precision/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/mixture-of-experts/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/multimodality/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/normalization/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/parallelism/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/positional-encoding/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/ppo/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/pretraining-objectives/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/qlora/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/rag/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/reasoning-models/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/regularization/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/retrieval-embeddings/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/reward-models/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/rlvr/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/rope/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/safety-and-refusal/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/scaling-laws/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/softmax/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/sparse-attention/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/speculative-decoding/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/state-space-models/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/superposition/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/supervised-fine-tuning/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/test-time-compute/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/residual-stream/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/transformer-block/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/tokenization/2026-08-23T00:00:00+00:00https://www.fanfani.net/articles/tool-use/2026-08-23T00:00:00+00:00https://www.fanfani.net/tutorials/2026-06-30T22:18:15+02:00https://www.fanfani.net/posts/2026-04-09T12:34:53+01:00https://www.fanfani.net/posts/09-april-2026/2026-04-09T12:34:53+01:00https://www.fanfani.net/posts/16-february-2026/2026-02-16T12:34:53+01:00https://www.fanfani.net/posts/07-february-2026/2026-02-07T12:34:53+01:00https://www.fanfani.net/posts/02-february-2026/2026-02-02T12:34:53+01:00https://www.fanfani.net/posts/01-february-2026/2026-02-01T12:34:53+01:00https://www.fanfani.net/posts/21-january-2026/2026-01-25T12:34:53+01:00https://www.fanfani.net/posts/20-january-2026/2026-01-20T12:34:53+01:00https://www.fanfani.net/posts/19-january-2026/2026-01-19T12:34:53+01:00https://www.fanfani.net/posts/10-january-2026/2026-01-10T12:34:53+01:00https://www.fanfani.net/posts/03-january-2026/2026-01-03T12:34:53+01:00https://www.fanfani.net/posts/27-december-2025/2025-12-27T12:34:53+01:00https://www.fanfani.net/posts/26-december-2025/2025-12-26T12:34:53+01:00https://www.fanfani.net/posts/23-december-2025/2025-12-23T12:34:53+01:00https://www.fanfani.net/posts/12-december-2025/2025-12-12T12:34:53+01:00https://www.fanfani.net/posts/07-december-2025/2025-12-07T21:34:53+01:00https://www.fanfani.net/posts/04-december-2025/2025-12-04T21:34:53+01:00https://www.fanfani.net/posts/02-november-2025/2025-11-02T21:34:53+01:00https://www.fanfani.net/posts/22-october-2025/2025-10-20T21:34:53+01:00https://www.fanfani.net/about/https://www.fanfani.net/links/https://www.fanfani.net/tutorials/moe-lessons/appendix-a-moe-family/https://www.fanfani.net/tutorials/moe-lessons/appendix-b-dense-vs-moe/https://www.fanfani.net/tutorials/moe-lessons/appendix-c-acronyms/https://www.fanfani.net/archives/https://www.fanfani.net/categories/https://www.fanfani.net/search/https://www.fanfani.net/tags/