payelb's picture
Upload PPO-aligned Llama-3.2-1B model using baseline DeBERTa reward model on UltraFeedback_openbmb
4fb5c1f verified
# payelb/UltraFeedback_openbmb_Llama-3.2-1B_aligned_with_baseline_deberta_RM
Base model: meta-llama/Llama-3.2-1B-Instruct
Alignment dataset: openbmb/UltraFeedback
Reward model: payelb/UltraFeedback_openbmb_reward-model-deberta-v3-base_1k_fixed_baseline
Method: PPO alignment with LoRA adapters.
Notes:
- Reward normalization and clipping enabled
- KL control enabled
- pad_token_id/eos_token_id explicitly set
- DeBERTa RM loaded on a single device (no device_map='auto')