| { | |
| "architectures": [ | |
| "MiniMindVLM" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "model_vlm.VLMConfig", | |
| "AutoModelForCausalLM": "model_vlm.MiniMindVLM" | |
| }, | |
| "bos_token_id": 1, | |
| "dropout": 0.0, | |
| "dtype": "bfloat16", | |
| "eos_token_id": 2, | |
| "flash_attn": true, | |
| "head_dim": 96, | |
| "hidden_act": "silu", | |
| "hidden_size": 768, | |
| "image_hidden_size": 768, | |
| "image_ids": [ | |
| 12 | |
| ], | |
| "image_special_token": "<|image_pad|>", | |
| "image_token_len": 64, | |
| "inference_rope_scaling": false, | |
| "intermediate_size": 2432, | |
| "max_position_embeddings": 32768, | |
| "max_seq_len": 8192, | |
| "model_type": "minimind-v", | |
| "moe_intermediate_size": 2432, | |
| "norm_topk_prob": true, | |
| "num_attention_heads": 8, | |
| "num_experts": 4, | |
| "num_experts_per_tok": 1, | |
| "num_hidden_layers": 8, | |
| "num_key_value_heads": 4, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 1000000.0, | |
| "router_aux_loss_coef": 0.0005, | |
| "transformers_version": "4.57.6", | |
| "use_moe": false, | |
| "vocab_size": 6400 | |
| } | |