# Learning coverage map — every subsystem has a home This map is the guarantee that the learning corpus **covers everything in the project**. The ground-up [`curriculum.md`](curriculum.md) is the _spine_ (10 modules, first principles to self-improving factory). This page is the _net_: every shipped capability in `PROJECT_STATUS.md` maps to a learning anchor here, so no subsystem is a black box with no explainer. How to read it: the **Anchor** column is where to go to learn that thing. Where a concept has a definitive external source, the anchor stays thin and links out (per the repo's "lean on external sources" rule) rather than re-explaining. The spine teaches you to _read_ the whole system; this map is the index that proves nothing was left unread. The runnable lab graph lives in [`path-registry.json`](path-registry.json). It gives every path prerequisites, an ordered sequence, repository anchors, a hands-on lab, a mastery gate, and validated CLI entry points. The public `/learn` cards also resolve the recipes assigned to each path and expose every evidence anchor rather than only the first guide. The public walkthrough at [`/learn`](https://posttrainllm.com/learn) renders the same paths, while [`../recipes/registry.json`](../recipes/registry.json) maps every retained technique recipe back to one of them. --- ## 1. Foundations — the ground-up spine The 10-module arc. This is the "learn it in order" path; everything below builds on it. | Module | Topic | Anchor | | -----: | ------------------------------------------- | ------------------------------------------------------------------------------------------------------------ | | 1 | Functions, data, parameters | [`session-01-neural-net-basics.md`](session-01-neural-net-basics.md) | | 2 | Loss and gradient descent | [`session-02-gradient-descent.md`](session-02-gradient-descent.md) | | 3 | Vectors, matrices, tensors | [`session-09-tensors.md`](session-09-tensors.md) | | 4 | Non-linear nets + backprop | [`session-03-non-linearities.md`](session-03-non-linearities.md) | | 5 | ML paradigms and scaling | [`session-04-ml-paradigms.md`](session-04-ml-paradigms.md), [`session-05-scaling.md`](session-05-scaling.md) | | 6 | Tokenization, embeddings, LM | [`session-06-tokenization-embeddings.md`](session-06-tokenization-embeddings.md) | | 7 | Attention and transformer blocks | [`session-10-attention.md`](session-10-attention.md) | | 8 | Training mechanics | [`session-08-training-mechanics.md`](session-08-training-mechanics.md) | | 9 | Post-training: SFT, LoRA, preference tuning | [`session-07-behavior-learning.md`](session-07-behavior-learning.md) | | 10 | Evals, rewards, self-improvement | [`session-11-evals-rewards.md`](session-11-evals-rewards.md) | Reference layer under the spine: [`llm-mechanics-fundamentals.md`](llm-mechanics-fundamentals.md) (RoPE, GQA, MoE, attention variants) and [`essential-vs-optimization.md`](essential-vs-optimization.md) (which math defines the model vs which only makes it fast). --- ## 2. Post-training methods and internals Everything under `PROJECT_STATUS.md` → "Training and post-training." | Capability | Anchor | | -------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | SFT | [`../training/sft.md`](../training/sft.md), Module 9 | | Pretraining | [`../training/pretrain.md`](../training/pretrain.md), Session 8 | | LoRA / DoRA / QLoRA | [`../lora_guide.md`](../techniques/lora_guide.md), [`../peft_variants.md`](../techniques/peft_variants.md), [`../factory/lora-geometry.md`](../factory/lora-geometry.md), Session 9 | | Encoder-decoder (seq2seq) adapters, copy bias, edit-aware loss | [`encoder-decoder-adapters.md`](encoder-decoder-adapters.md), [`../factory/autocorrect-adapter-recipe.md`](../factory/autocorrect-adapter-recipe.md) | | DPO / SimPO / preference tuning | [`../training/dpo.md`](../training/dpo.md), Module 9 + 10 | | Distillation | [`../distillation.md`](../techniques/distillation.md), [`diversity-driven-small-model-reasoning.md`](diversity-driven-small-model-reasoning.md) | | Evolution strategies (ES) | [`../evolution_strategies.md`](../techniques/evolution_strategies.md), [`castform-rl-finetune.md`](castform-rl-finetune.md) | | RLVR / ReST / GRPO | [`advanced-llm-training.md`](advanced-llm-training.md), [`../GRPO_CLARIFY.md`](../GRPO_CLARIFY.md) | | Optimizers, schedules, stability | [`../optimizers.md`](../techniques/optimizers.md), [`../galore_and_stability.md`](../techniques/galore_and_stability.md), Session 8 | | NEFTune, z-loss, WSD, LLRD, seq packing, grad checkpointing | [`../training_guide.md`](../guides/training_guide.md), [`../gradient_checkpointing_results.md`](../performance/gradient_checkpointing_results.md), Session 8 | | Precision (bf16 / fp8 / mixed) | [`../precision.md`](../techniques/precision.md), [`advanced-llm-training.md`](advanced-llm-training.md) | | Method-vs-recipe discipline | [`../techniques/method-vs-recipe.md`](../techniques/method-vs-recipe.md) | --- ## 3. Data pipeline `PROJECT_STATUS.md` → "Data." | Capability | Anchor | | ------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------ | | Dataset registry / inventory | [`../dataset-inventory.md`](../dataset-inventory.md), [`../data_inventory.md`](../data_inventory.md) | | HF integration | [`../hf_datasets_integration.md`](../integrations/hf_datasets_integration.md) | | GitHub fetcher | [`../github_data_integration.md`](../integrations/github_data_integration.md) | | Magpie / synthesis | [`advanced-llm-training.md`](advanced-llm-training.md) (data curation), [`small-model-tool-calling-playbook.md`](small-model-tool-calling-playbook.md) | | Tokenizer training, extractor data | [`session-06-tokenization-embeddings.md`](session-06-tokenization-embeddings.md), [`../tool_call_extractor.md`](../tool_call_extractor.md) | | Traces → data, corrections → data | [`../recipes/from-traces.md`](../recipes/from-traces.md) | | Quality filter, dedupe, reasoning-classify | [`../factory/post-training-factory.md`](../factory/post-training-factory.md), [`castform-rl-finetune.md`](castform-rl-finetune.md) | --- ## 4. Evaluation `PROJECT_STATUS.md` → "Evals." Module 10 is the concept spine; these are the surfaces. | Capability | Anchor | | -------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Eval protocol, frozen baselines, gates | [`../factory/eval-protocol.md`](../factory/eval-protocol.md), Module 10 | | Eval methodology / broken-eval lessons | [`eval-methodology-2026-06-08.md`](eval-methodology-2026-06-08.md), [`eval-matrix-2026-06-08.md`](eval-matrix-2026-06-08.md) | | BFCL / tool-calling eval | [`tool-calling-frontier-parity.md`](tool-calling-frontier-parity.md), [`small-model-tool-calling-playbook.md`](small-model-tool-calling-playbook.md) | | SQL execution / exact / slices / candidate-choice | [`../techniques/sql-technique-backlog.md`](../techniques/sql-technique-backlog.md), Module 10 | | lm-eval / HumanEval / MTEB / MILU / tau-bench | [`../lm_eval_integration.md`](../integrations/lm_eval_integration.md), [`advanced-ml-systems-eval.md`](advanced-ml-systems-eval.md) | | Router / routed-specialist eval, eval-gate, planner eval | [`../recipes/eval-gate.md`](../recipes/eval-gate.md), [`../recipes/eval_planner.md`](../recipes/eval_planner.md), [`../planner-lock-2026-06-19.md`](../sessions/planner-lock-2026-06-19.md) | | LLM-as-judge, perplexity, contamination | [`advanced-ml-systems-eval.md`](advanced-ml-systems-eval.md), Module 10 | | Leaderboard / reporting | [`../leaderboard.md`](../leaderboard.md), [`../factory/reports.md`](../factory/reports.md) | --- ## 5. Runtime, serving, and agents `PROJECT_STATUS.md` → "Runtime/serving." | Capability | Anchor | | -------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Inference/serving architecture, batching, roofline | [`advanced-llm-inference.md`](advanced-llm-inference.md) | | KV cache + paging | [`../kv_cache_optimization.md`](../performance/kv_cache_optimization.md), [`advanced-llm-inference.md`](advanced-llm-inference.md) | | OpenAI/Ollama-compatible serve, Continue provider | [`../agent_runtime.md`](../agent_runtime.md), [`../continue_provider.md`](../integrations/continue_provider.md) | | Agent loop, tool dispatch | [`model-vs-agent.md`](model-vs-agent.md), [`../async_tool_dispatch.md`](../async_tool_dispatch.md), [`agent-context-hierarchy.md`](agent-context-hierarchy.md) | | Constrained JSON / FSM generation | [`../constrained_generation.md`](../techniques/constrained_generation.md) | | Speculative decoding / MTP | [`../speculative_heads.md`](../techniques/speculative_heads.md), [`../mtp.md`](../techniques/mtp.md), [`advanced-llm-inference.md`](advanced-llm-inference.md) | | Cost routing / escalation / cascade (AutoMix, ScaleDown) | [`../recipes/automix.md`](../recipes/automix.md), [`../recipes/b25-scaledown.md`](../recipes/b25-scaledown.md), [`agent-context-hierarchy.md`](agent-context-hierarchy.md) | | Streaming / long context (StreamingLLM, KIVI) | [`../streaming_llm_kivi.md`](../techniques/streaming_llm_kivi.md) | --- ## 6. Quantization, packaging, and inference optimization `PROJECT_STATUS.md` → "Packaging/runtime." | Capability | Anchor | | ---------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------- | | Quantization (GGUF / AWQ / GPTQ / HQQ), quant theory | [`../quantization_expansion.md`](../techniques/quantization_expansion.md), [`advanced-llm-inference.md`](advanced-llm-inference.md) | | Export to MLX / safetensors / CoreML | [`../recipes/mlx-export.md`](../recipes/mlx-export.md), [`../factory/packaging.md`](../factory/packaging.md) | | merge / bake-lora | [`../factory/lora-geometry.md`](../factory/lora-geometry.md), [`../lora_guide.md`](../techniques/lora_guide.md) | | Pruning | [`../pruning.md`](../techniques/pruning.md) | | Specialist packaging / model cards | [`../factory/packaging.md`](../factory/packaging.md), [`../factory/public-artifacts.md`](../factory/public-artifacts.md) | --- ## 7. Attention and kernel internals The math oracle is [`python_ref/model.py`](../../python_ref/model.py); Session 10 is the concept. | Capability | Anchor | | --------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------ | | Attention math, transformer block | [`session-10-attention.md`](session-10-attention.md) | | FlashAttention-2 forward/backward | [`../fa2_forward_notes.md`](../performance/fa2_forward_notes.md), [`../fa2_backward_notes.md`](../performance/fa2_backward_notes.md) | | Online softmax | [`../online_softmax_in_attention.md`](../performance/online_softmax_in_attention.md) | | MoE / expert routing | [`../moe.md`](../techniques/moe.md), [`llm-mechanics-fundamentals.md`](llm-mechanics-fundamentals.md) | --- ## 8. Browser / WASM / WebGPU track (completed, parked) `PROJECT_STATUS.md` → "Completed/parked learning tracks." | Capability | Anchor | | ------------------------------------------------------ | ----------------------------------------------------------------------------------------------------- | | WebGPU execution model (read before the `.wgsl` files) | [`webgpu-execution-model.md`](webgpu-execution-model.md) | | Browser GPT training, WASM SIMD, OPFS | [`../browser_notes.md`](../browser_notes.md) | | BPE-in-browser scoring | [`../bpe_browser_scoring.md`](../bpe_browser_scoring.md) | | Numerics gates / precision drift | [`../precision.md`](../techniques/precision.md), [`../determinism.md`](../performance/determinism.md) | --- ## 9. Native Mac / MLX / ANE / Apple Foundation Models | Capability | Anchor | | -------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------- | | Mac-local mastery map (living agenda) | [`mac-mastery-map.md`](mac-mastery-map.md) | | Apple on-device Foundation Models — where they fit | [`apple-on-device-foundation-models.md`](apple-on-device-foundation-models.md), [`app-intents-comparison.md`](app-intents-comparison.md) | | ANE / CoreML research (negative results) | [`ane-research/`](ane-research/dossier) | | Native runtime architecture | [`../../native-mac/ARCHITECTURE.md`](https://github.com/PostTrainLLM/posttrainllm/blob/main/native-mac/ARCHITECTURE.md) | --- ## 10. Interpretability (completed, parked) | Capability | Anchor | | ------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------- | | Attention heatmap + logit lens (browser) | [`../interpretability.md`](../techniques/interpretability.md), Session 10 | | SAE, ROME, MEMIT, tuned/logit lens, activation patching | [`../interpretability.md`](../techniques/interpretability.md), [`advanced-ml-systems-eval.md`](advanced-ml-systems-eval.md) | --- ## 11. Multimodal (VLM) — research/parked | Capability | Anchor | | ----------------------------------------------------- | ------------------------------------------------------------ | | Qwen3-VL mRoPE + DeepStack, vision-language attention | [`qwen3-vl-mrope-deepstack.md`](qwen3-vl-mrope-deepstack.md) | --- ## 12. The factory loop and methodology The thing all of the above serves. | Capability | Anchor | | -------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------- | | Factory overview + run schema | [`../factory/overview.md`](../factory/overview.md), [`../factory/run-schema.md`](../factory/run-schema.md) | | Batch post-training / rollout plan | [`../factory/batch-posttraining.md`](../factory/batch-posttraining.md) | | Case-study reports, public artifacts | [`../factory/case-study-template.md`](../factory/case-study-template.md), [`../factory/public-artifacts.md`](../factory/public-artifacts.md) | | Attempt history (worked/failed/regressed) | [`../attempt-ledger.md`](../attempt-ledger.md) | | Owner learning sequence tied to factory work | [`../learning-pipeline.md`](../learning-pipeline.md), [`../learning-progress.md`](../learning-progress.md) | --- ## 13. Strategy and external knowledge | Capability | Anchor | | ---------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- | | Competitive landscape, Mac-first whitespace | [`competitive-landscape.md`](competitive-landscape.md) | | Reviewed external products / stolen techniques | [`../external-products-reviewed.md`](../external-products-reviewed.md), [`castform-rl-finetune.md`](castform-rl-finetune.md) | | Speech & systems interview-grade topics | [`speech-and-systems-topics.md`](speech-and-systems-topics.md) | | Papers / reading list | [`external-references.md`](external-references.md), [`../CITATIONS.md`](../CITATIONS.md) | | Running questions and tangents | [`journal.md`](journal.md) | --- ## Coverage Guarantee **Every shipped capability in `PROJECT_STATUS.md` → "Features (shipped)" and "Completed/parked learning tracks" appears in exactly one section above with a learning anchor.** The factory-loop framing is the checklist: ```text target -> data -> post-training -> eval -> package -> report ``` - **target / methodology** → §12 - **data** → §3 - **post-training** → §2, §7 (internals), §9 (Mac runtime), §11 (VLM) - **eval** → §4, Module 10 - **package** → §6 - **report** → §4, §12 - **serve/run the result** → §5 - **understand the result** → §10 (interpretability), §1 (foundations) ### Closed-lab coverage rule This map is guarded by [`../../scripts/docs-checks/check_learning_roadmap.py`](../../scripts/docs-checks/check_learning_roadmap.py) (run via `bash evals/learning-roadmap-smoke.sh`). When a **new subsystem** ships in `PROJECT_STATUS.md`, add a row here pointing to its best explainer before calling the feature done — a capability with no learning anchor is an undocumented black box. When an existing doc is the definitive explainer, point to it (DRY); only write a new learn doc when nothing adequate exists. For this closed historical project, the validator additionally requires every recipe ID to resolve to one of the path IDs in `path-registry.json`. Fresh experiments may extend the graph only after the owner starts a new question.