Accueil Explorer Aide
S'inscrire Connexion
lainlives
/
sft-trainer
1
0
Fork 0
Fichiers Tickets 0 Pull Requests 0 Wiki

Aucune description

Aborescence: 1bbc0df33e
Branches Tags
sft-trainer
ZIP TAR.GZ
Sameric 1bbc0df33e Upload Dockerfile: expandable_segments fix + reduced params il y a 4 mois
scripts 93538163f5 Deploy Qwen3-Coder-Next uncensored fine-tuner il y a 4 mois
.gitattributes 1cf7fc58ca initial commit il y a 4 mois
.gitignore 712316c058 Deploy Qwen3-Coder-Next uncensored fine-tuner il y a 4 mois
Dockerfile 1bbc0df33e Upload Dockerfile: expandable_segments fix + reduced params il y a 4 mois
README.md 712316c058 Deploy Qwen3-Coder-Next uncensored fine-tuner il y a 4 mois
app.py 67fe1cd68a fix: match UI defaults to reduced memory params il y a 4 mois
config.yaml 494d4838fd Deploy Qwen3-Coder-Next uncensored fine-tuner il y a 4 mois
requirements.txt 494d4838fd Deploy Qwen3-Coder-Next uncensored fine-tuner il y a 4 mois
train.py bac3cf12ad fix: aggressive OOM prevention - lora_r=16, seq_len=512, max_samples=5000, clear cache il y a 4 mois

README.md

title: Qwen3-Coder-Next Uncensored Fine-Tuning emoji: "\U0001F525" colorFrom: red colorTo: yellow sdk: docker app_port: 7860 suggested_hardware: a100-large suggested_storage: large pinned: true

license: apache-2.0

Qwen3-Coder-Next Uncensored Fine-Tuner

Fine-tune Qwen/Qwen3-Coder-Next (80B MoE / 3B active) using QLoRA to remove refusal behavior.

Setup

  1. Create a HF Space with Docker SDK and a100-large hardware
  2. Add Secrets in Space Settings:
    • HF_TOKEN — your Hugging Face write token
    • WANDB_API_KEY — (optional) for training metrics logging
  3. Upload this repo to the Space
  4. The Gradio UI will let you configure and launch training

Features

  • QLoRA 4-bit fine-tuning (fits on single A100 80GB)
  • Multiple uncensored dataset options built-in
  • Custom dataset upload support
  • Real-time training metrics in UI
  • One-click LoRA merge + push to Hub
  • Abliteration support (no training needed)

Hardware Requirements

Method VRAM Recommended HF Hardware
QLoRA 4-bit ~45-60GB A100 Large (80GB)
Abliteration ~45-60GB A100 Large (80GB)