SEGAgentRL/LLDS-A-GRPO-Qwen2.5-3B-Base-MA
Partial fine-tune vs Qwen/Qwen2.5-3B. Partial change: 90% of the weights changed (attention, mlp, norms), in 36 of 36 layers.
Tensors changed: Embeddings 0/1, Norms 63/73, MLP 108/108, Attention 252/252
2 derivatives checked
- mradermacher/LLDS-A-GRPO-Qwen2.5-3B-Base-MA-i1-GGUF: Quantization
- mradermacher/LLDS-A-GRPO-Qwen2.5-3B-Base-MA-GGUF: Same weights, converted