xrsula/MNLP_M3_mcqa_model

1{2  "architectures": [3    "Qwen3ForCausalLM"4  ],5  "attention_bias": false,6  "attention_dropout": 0.0,7  "eos_token_id": 151643,8  "head_dim": 128,9  "hidden_act": "silu",10  "hidden_size": 1024,11  "initializer_range": 0.02,12  "intermediate_size": 3072,13  "max_position_embeddings": 32768,14  "max_window_layers": 28,15  "model_type": "qwen3",16  "num_attention_heads": 16,17  "num_hidden_layers": 28,18  "num_key_value_heads": 8,19  "pad_token_id": 151654,20  "rms_norm_eps": 1e-06,21  "rope_scaling": null,22  "rope_theta": 1000000,23  "sliding_window": null,24  "tie_word_embeddings": true,25  "torch_dtype": "bfloat16",26  "transformers_version": "4.51.3",27  "unsloth_fixed": true,28  "unsloth_version": "2025.5.7",29  "use_cache": true,30  "use_sliding_window": false,31  "vocab_size": 15193632}33