A straightforward method for training your LLM, from downloading data to generating text.
Add link to Pretraining at bottom.
Mr. Ohu2b59fac7fd8c
diff --git a/docs/01_data_pipeline.md b/docs/01_data_pipeline.mdindex b14874c..403acc2 100644--- a/docs/01_data_pipeline.md+++ b/docs/01_data_pipeline.md@@ -157,3 +157,6 @@ as the gold it compares against.| `preferences.jsonl` (+ `_test`) | `{prompt, chosen, rejected}` | Reward Model, DPO || `rl_prompts_train.jsonl` / `_test` | `{prompt, gold}` | PPO, GRPO || `arithmetic_prompts.jsonl` | `{prompt, gold}` | GRPO curriculum warm-up |+<br>++➡️ Next: [Stage 1 — Pretraining](02_pretraining.md).