haddock-development/ttc-pi-package/ttc-training

Claude Code and agent skill by haddock-development/ttc-pi-package

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

TTC-aware training skill for Qwen models. Use when training models with Test-Time Compute awareness, including CPT → SFT → RLVR → TTC pipeline stages. Supports Kaggle and HF Jobs backends.

Install

npx vskill@latest install haddock-development/ttc-pi-package/ttc-training

SKILL.md

TTC Training Skill

Training pipeline for Qwen models with TTC (Test-Time Compute) awareness.

Setup

bash
# Ensure kaggle CLI is configured
kaggle competitions list
 
# Or for HF Jobs
huggingface-cli login

Training Stages

1. CPT (Continued Pre-Training)
bash
# From ttc-inference-scaling-stack repo
python scripts/cpt_train.py \
  --model Qwen/Qwen2.5-0.5B \
  --dataset path/to/corpus \
  --backend kaggle \
  --output-dir models/cpt-qwen
2. SFT (Supervised Fine-Tuning)
bash
python scripts/sft_train.py \
  --base-model models/cpt-qwen \
  --dataset path/to/sft_data \
  --lora-r 16 \
  --backend kaggle
3. RLVR (Reinforcement Learning with Verifiable Rewards)
bash
python scripts/rlvr_train.py \
  --sft-model models/sft-qwen \
  --verifier qwen-verifier \
  --backend hf-jobs
4. TTC-Aware Training
bash
python scripts/ttc_train.py \
  --rlvr-model models/rlvr-qwen \
  --router-profile balanced \
  --backend hf-jobs

Profiles

ProfileComputeUse Case
low~1 GPU-hourQuick experiments, smoke tests
medium~5 GPU-hoursStandard training, Kaggle
high~20+ GPU-hoursProduction runs, HF Jobs

Backends

  • kaggle: Free T4/P100, 30h/week limit
  • hf-jobs: Paid A10G/L4, unlimited
  • local: Your own GPU

Expected Outputs

After training completion:

  • models/<stage>-qwen/final_model/ - Model weights
  • models/<stage>-qwen/train_summary.json - Training metrics
  • models/<stage>-qwen/config.json - Training configuration

Example

User: Train Qwen 3.5 0.8B on my instruction dataset with SFT, use Kaggle backend
 
Agent will:
1. Load skill instructions (this file)
2. Prepare Kaggle notebook configuration
3. Launch training job
4. Monitor and report results
5. Save model to specified location

References

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

See the full security report

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

See evaluation results

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/haddock-development/ttc-pi-package/ttc-training/badge)](https://verified-skill.com/skills/haddock-development/ttc-pi-package/ttc-training)

Something wrong with this skill or its page? Tell us.