Add SWIFT coding agent probe experiment scripts

2026-06-24 21:55:23 +08:00
commit 73e9752331
19 changed files with 526 additions and 0 deletions
@@ -0,0 +1,81 @@
+# TI Coding Agent Probe Skill
+
+Use this skill when the user wants to run or modify the TI coding-agent SFT probe experiments based on the Hugging Face dataset `ti_coding_agent_training_probe_20260624`.
+
+## Repository Contract
+
+- Work in `/ssd/workspace/yi/ti_coding_agent_probe` on B300 unless the user says otherwise.
+- Use the B300 proxy for all network access:
+  - `http_proxy=http://100.72.0.101:8888`
+  - `https_proxy=http://100.72.0.101:8888`
+  - `HF_ENDPOINT=https://hf-mirror.com`
+- Do not install Python packages globally. Use the repository-local `.venv` created by `scripts/setup_env.sh`.
+- Do not start GPU training before checking GPU occupancy with `nvidia-smi`.
+- Do not commit or upload `data/`, `models/`, `outputs/`, `runs/`, or `logs/`.
+
+## Key Commands
+
+Initialize the repo and environment:
+
+```bash
+git submodule update --init --recursive
+./scripts/setup_env.sh
+```
+
+Download the training probe dataset:
+
+```bash
+export HF_ENDPOINT=https://hf-mirror.com
+export HF_DATASET_REPO_ID=<owner>/ti_coding_agent_training_probe_20260624
+./scripts/download_dataset.py
+```
+
+Download base models:
+
+```bash
+./scripts/download_models.sh
+```
+
+Run the full ordered experiment:
+
+```bash
+./scripts/run_all_experiments.sh
+```
+
+Run only one stage:
+
+```bash
+./scripts/train_qwen35_9b_lora.sh
+./scripts/train_qwen35_9b_full.sh
+./scripts/train_qwen36_27b_lora.sh
+./scripts/train_qwen36_27b_full.sh
+```
+
+Open TensorBoard:
+
+```bash
+./scripts/tensorboard.sh
+```
+
+## Training Semantics
+
+The dataset uses SWIFT-style chat messages. `system`, `user`, and `tool` messages should remain masked with `loss=false`; only assistant trajectories should contribute to loss. This keeps scaffold prompts and tool outputs as conditioning context rather than targets to memorize.
+
+The default experiment uses:
+
+- 1 epoch
+- LoRA rank 32 for LoRA runs
+- bf16 full fine-tuning for full runs
+- `max_length=262144`
+- checkpoint save every 1000 steps
+- validation every 1000 steps
+- TensorBoard logging under `runs/`
+
+Override model IDs or paths with:
+
+```bash
+export QWEN35_9B_MODEL_ID=<hf-id>
+export QWEN36_27B_MODEL_ID=<hf-id>
+export QWEN35_9B_MODEL_PATH=<local-path>
+export QWEN36_27B_MODEL_PATH=<local-path>
+```