https://llmstudy.shddai.net/zh/2026-08-09 https://llmstudy.shddai.net/en/2026-08-09 https://llmstudy.shddai.net/zh/world-models/2026-08-09 https://llmstudy.shddai.net/en/world-models/2026-08-09 https://llmstudy.shddai.net/zh/lesson/p-1-python-from-zero-expressions-to-testable-functions/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/p-1-python-from-zero-expressions-to-testable-functions/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/p-2-python-engineering-habits-environments-files-exceptions-and-pytest/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/p-2-python-engineering-habits-environments-files-exceptions-and-pytest/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/p-3-from-numpy-arrays-to-pytorch-tensors/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/p-3-from-numpy-arrays-to-pytorch-tensors/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/p-4-prerequisite-gate-run-your-first-training-loop/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/p-4-prerequisite-gate-run-your-first-training-loop/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/0-1-map-the-llm-stack-and-how-to-learn-it/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/0-1-map-the-llm-stack-and-how-to-learn-it/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/0-2-python-objects-iterators-and-vectorized-thinking/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/0-2-python-objects-iterators-and-vectorized-thinking/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/0-3-the-linear-algebra-models-actually-use/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/0-3-the-linear-algebra-models-actually-use/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/0-4-probability-information-and-cross-entropy/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/0-4-probability-information-and-cross-entropy/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/0-5-derivatives-partials-and-the-chain-rule/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/0-5-derivatives-partials-and-the-chain-rule/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/0-6-pytorch-tensors-and-reproducible-experiments/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/0-6-pytorch-tensors-and-reproducible-experiments/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/0-7-mastery-gate-0-read-a-training-loop/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/0-7-mastery-gate-0-read-a-training-loop/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/1-1-a-neuron-is-not-magic-functions-parameters-activations/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/1-1-a-neuron-is-not-magic-functions-parameters-activations/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/1-2-the-chain-rule-is-a-message-passing-rule/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/1-2-the-chain-rule-is-a-message-passing-rule/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/1-3-make-gradients-flow-backward-through-a-graph/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/1-3-make-gradients-flow-backward-through-a-graph/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/1-4-gradient-checking-verify-before-trusting-code/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/1-4-gradient-checking-verify-before-trusting-code/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/1-5-from-value-to-neuron-layer-and-mlp/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/1-5-from-value-to-neuron-layer-and-mlp/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/1-6-losses-regularization-and-sgd/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/1-6-losses-regularization-and-sgd/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/1-7-why-neural-networks-fail-to-train/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/1-7-why-neural-networks-fail-to-train/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/1-8-mastery-gate-1-rewrite-micrograd-closed-book/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/1-8-mastery-gate-1-rewrite-micrograd-closed-book/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-1-bigram-the-smallest-language-model/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-1-bigram-the-smallest-language-model/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-2-negative-log-likelihood-the-model-scorecard/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-2-negative-log-likelihood-the-model-scorecard/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-3-embeddings-place-discrete-symbols-in-space/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-3-embeddings-place-discrete-symbols-in-space/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-4-mlp-language-models-and-data-splits/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-4-mlp-language-models-and-data-splits/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-5-training-deep-nets-activations-gradients-batchnorm/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-5-training-deep-nets-activations-gradients-batchnorm/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-6-manual-backprop-become-a-backprop-ninja/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-6-manual-backprop-become-a-backprop-ninja/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-7-bpe-from-bytes-to-subwords/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-7-bpe-from-bytes-to-subwords/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-8-the-hidden-cost-of-tokenization/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-8-the-hidden-cost-of-tokenization/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/2-9-mastery-gate-2-build-minbpe-makemore/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/2-9-mastery-gate-2-build-minbpe-makemore/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-1-context-aggregation-from-averaging-to-attention/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-1-context-aggregation-from-averaging-to-attention/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-2-scaled-dot-product-attention/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-2-scaled-dot-product-attention/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-3-multi-head-attention-read-different-relations-in-parallel/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-3-multi-head-attention-read-different-relations-in-parallel/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-4-position-residual-streams-and-normalization/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-4-position-residual-streams-and-normalization/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-5-implement-a-transformer-block-line-by-line/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-5-implement-a-transformer-block-line-by-line/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-6-gpt-autoregressive-objectives-and-weight-tying/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-6-gpt-autoregressive-objectives-and-weight-tying/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-7-sampling-is-not-a-random-button/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-7-sampling-is-not-a-random-button/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-8-read-gpt-2-configuration-parameter-counts-weight-loading/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-8-read-gpt-2-configuration-parameter-counts-weight-loading/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-9-reproduce-124m-from-correct-to-fast/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-9-reproduce-124m-from-correct-to-fast/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/3-10-mastery-gate-3-draw-and-implement-gpt-closed-book/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/3-10-mastery-gate-3-draw-and-implement-gpt-closed-book/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-1-budget-before-training-parameters-flops-memory/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-1-budget-before-training-parameters-flops-memory/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-2-data-engineering-collect-clean-deduplicate-mix/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-2-data-engineering-collect-clean-deduplicate-mix/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-3-every-control-in-the-training-loop/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-3-every-control-in-the-training-loop/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-4-mixed-precision-and-numerical-stability/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-4-mixed-precision-and-numerical-stability/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-5-gpus-kernels-and-the-memory-hierarchy/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-5-gpus-kernels-and-the-memory-hierarchy/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-6-flashattention-and-triton-fundamentals/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-6-flashattention-and-triton-fundamentals/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-7-distributed-training-dp-tp-pp-fsdp/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-7-distributed-training-dp-tp-pp-fsdp/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-8-scaling-laws-and-chinchilla-intuition/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-8-scaling-laws-and-chinchilla-intuition/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-9-training-observability-and-recovery/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-9-training-observability-and-recovery/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/4-10-mastery-gate-4-from-nanogpt-to-small-pretraining/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/4-10-mastery-gate-4-from-nanogpt-to-small-pretraining/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-1-the-gap-between-a-base-model-and-an-assistant/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-1-the-gap-between-a-base-model-and-an-assistant/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-2-sft-rewrite-behavioral-priors-with-demonstrations/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-2-sft-rewrite-behavioral-priors-with-demonstrations/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-3-preference-data-and-reward-models/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-3-preference-data-and-reward-models/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-4-dpo-preference-optimization-without-a-reward-model/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-4-dpo-preference-optimization-without-a-reward-model/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-5-ppo-grpo-and-rl-post-training/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-5-ppo-grpo-and-rl-post-training/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-6-evaluation-define-good-before-optimizing/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-6-evaluation-define-good-before-optimizing/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-7-hallucination-calibration-and-uncertainty/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-7-hallucination-calibration-and-uncertainty/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-8-safety-red-teaming-and-system-cards/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-8-safety-red-teaming-and-system-cards/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-9-look-inside-activations-an-interpretability-entry-point/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-9-look-inside-activations-an-interpretability-entry-point/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/5-10-mastery-gate-5-defend-a-post-training-plan/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/5-10-mastery-gate-5-defend-a-post-training-plan/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/6-1-prefill-and-decode-two-different-workloads/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/6-1-prefill-and-decode-two-different-workloads/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/6-2-kv-cache-trade-memory-for-repeated-compute/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/6-2-kv-cache-trade-memory-for-repeated-compute/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/6-3-quantization-numbers-error-calibration/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/6-3-quantization-numbers-error-calibration/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/6-4-llama-cpp-dissect-local-inference-engineering/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/6-4-llama-cpp-dissect-local-inference-engineering/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/6-5-vllm-pagedattention-and-continuous-batching/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/6-5-vllm-pagedattention-and-continuous-batching/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/6-6-service-reliability-limits-caching-degradation-observability/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/6-6-service-reliability-limits-caching-degradation-observability/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/6-7-mastery-gate-6-two-deployments-one-benchmark/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/6-7-mastery-gate-6-two-deployments-one-benchmark/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/7-1-the-minimal-agent-loop/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/7-1-the-minimal-agent-loop/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/7-2-tool-calling-and-structured-output/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/7-2-tool-calling-and-structured-output/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/7-3-context-engineering-and-memory/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/7-3-context-engineering-and-memory/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/7-4-planning-reflection-and-multi-agent-systems/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/7-4-planning-reflection-and-multi-agent-systems/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/7-5-agent-evaluation-from-final-answers-to-trajectories/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/7-5-agent-evaluation-from-final-answers-to-trajectories/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/7-6-production-boundaries-permissions-safety-human-oversight/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/7-6-production-boundaries-permissions-safety-human-oversight/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/7-7-nanochat-understand-a-complete-chat-system/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/7-7-nanochat-understand-a-complete-chat-system/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/7-8-mastery-gate-7-final-defense/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/7-8-mastery-gate-7-final-defense/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/8-1-moe-mla-and-multi-token-prediction/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/8-1-moe-mla-and-multi-token-prediction/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/8-2-reasoning-models-and-reinforcement-learning-with-verifiable-rewards/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/8-2-reasoning-models-and-reinforcement-learning-with-verifiable-rewards/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/8-3-long-context-is-not-one-number/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/8-3-long-context-is-not-one-number/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/8-4-dspark-speculative-generation-and-scheduling/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/8-4-dspark-speculative-generation-and-scheduling/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/8-5-diffusion-and-block-parallel-language-models/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/8-5-diffusion-and-block-parallel-language-models/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/8-6-post-training-for-agents/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/8-6-post-training-for-agents/ 2026-08-09 https://llmstudy.shddai.net/zh/lesson/8-7-kimi-k3-audit-a-3t-class-open-weight-system/ 2026-08-09 https://llmstudy.shddai.net/en/lesson/8-7-kimi-k3-audit-a-3t-class-open-weight-system/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-0-1-what-is-a-world-model-exactly/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-0-1-what-is-a-world-model-exactly/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-0-2-pomdps-modeling-a-world-you-cannot-fully-observe/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-0-2-pomdps-modeling-a-world-you-cannot-fully-observe/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-0-3-build-the-smallest-working-world-model/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-0-3-build-the-smallest-working-world-model/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-1-1-classic-world-models-vae-rnn-controller/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-1-1-classic-world-models-vae-rnn-controller/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-1-2-dreamer-train-an-agent-inside-latent-imagination/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-1-2-dreamer-train-an-agent-inside-latent-imagination/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-1-3-muzero-learn-only-the-dynamics-planning-needs/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-1-3-muzero-learn-only-the-dynamics-planning-needs/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-2-1-jepa-why-not-predict-pixels-directly/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-2-1-jepa-why-not-predict-pixels-directly/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-2-2-v-jepa-2-from-video-understanding-to-robot-planning/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-2-2-v-jepa-2-from-video-understanding-to-robot-planning/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-3-1-genie-1-3-from-latent-actions-to-real-time-worlds/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-3-1-genie-1-3-from-latent-actions-to-real-time-worlds/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-3-2-fei-fei-li-marble-and-spatial-intelligence/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-3-2-fei-fei-li-marble-and-spatial-intelligence/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-4-1-nvidia-cosmos-and-world-foundation-models/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-4-1-nvidia-cosmos-and-world-foundation-models/ 2026-08-09 https://llmstudy.shddai.net/zh/world-models/lesson/wm-4-2-evaluate-world-models-from-impressive-to-useful/ 2026-08-09 https://llmstudy.shddai.net/en/world-models/lesson/wm-4-2-evaluate-world-models-from-impressive-to-useful/ 2026-08-09