Documentation Index
Fetch the complete documentation index at: https://www.radixark.com/llms.txt
Use this file to discover all available pages before exploring further.
| Page | What it covers |
|---|---|
| Core Concepts | The four objects in the training loop and the four-knob invariant. |
| Argument Groups | Where MODEL_ARGS, PERF_ARGS, GRPO_ARGS, and the other launch-script arrays belong. |
| Training Backend | Megatron-LM as the training backend — parallelism, checkpoints, and hooks. |
| Training Script Walkthrough | The eight XXX_ARGS arrays in a launch script, plus the execution modes (sync/async, colocation, dynamic sampling, partial rollout, BF16+FP8). |
| Monitoring & Logging | wandb, structured logs, per-source breakdowns, profiling, router metrics. |
| Customization | The 22 --*-path plug-points for custom Python — rollout, reward, filters, loss, hooks. |
| Rollout Endpoints | The /generate endpoint and the OpenAI chat endpoint for agentic sessions. |
| Fully Async Rollout | Queue-backed rollout production, tuning knobs, and when to use train_async.py. |
| Agentic Chat Templates | Verifying and fixing the chat template so multi-turn rollout stays append-only. |
| CLI Reference | Every flag Miles accepts, grouped by subsystem. |
Which pages do I actually need?
- Training my first job — read Core Concepts, then Training Script Walkthrough.
- Tuning a running job — Training Script Walkthrough in depth + CLI Reference.
- Plugging in a custom reward / rollout / filter — skim Core Concepts for vocabulary, then go to Customization.
- Contributor onboarding — read top to bottom.