mxcui/pcgrad-imdb-ppo-prop0.2-alpha1.0-seed42-mean_kl0.1-Qwen-Qwen3-4B-Base

Re-upload merged full Qwen3-4B model; remove stale empty LoRA adapter

End of training

Upload tokenizer

Upload mxcui/pcgrad-imdb-ppo-prop0.2-alpha1.0-seed42-mean_kl0.1-Qwen-Qwen3-4B-Base model weights

initial commit