| { |
| "_name_or_path": "MubitLabs/banyan-sapling", |
| "architectures": [ |
| "BanyanSaplingVerifier" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 151643, |
| "chunk_size_feed_forward": 0, |
| "decoder_sparse_step": 1, |
| "dtype": "bfloat16", |
| "eos_token_id": 151645, |
| "head_dim": 128, |
| "hidden_act": "silu", |
| "hidden_size": 2048, |
| "id2label": { |
| "0": "supported", |
| "1": "violation", |
| "2": "insufficient_information", |
| "3": "inconsistent_context" |
| }, |
| "initializer_range": 0.02, |
| "intermediate_size": 6144, |
| "is_encoder_decoder": false, |
| "label2id": { |
| "supported": 0, |
| "violation": 1, |
| "insufficient_information": 2, |
| "inconsistent_context": 3 |
| }, |
| "max_position_embeddings": 262144, |
| "max_window_layers": 48, |
| "mlp_only_layers": [], |
| "model_type": "qwen3_moe", |
| "moe_intermediate_size": 768, |
| "norm_topk_prob": true, |
| "num_attention_heads": 32, |
| "num_experts_per_tok": 8, |
| "num_hidden_layers": 48, |
| "num_key_value_heads": 4, |
| "num_local_experts": 96, |
| "output_attentions": false, |
| "output_hidden_states": false, |
| "output_router_logits": false, |
| "pad_token_id": null, |
| "problem_type": null, |
| "return_dict": true, |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "rope_theta": 10000000, |
| "rope_type": "default" |
| }, |
| "router_aux_loss_coef": 0.001, |
| "sliding_window": null, |
| "tie_word_embeddings": false, |
| "transformers_version": "5.17.0", |
| "use_cache": true, |
| "use_sliding_window": false, |
| "vocab_size": 151936 |
| } |
|
|
