| { |
| "model_type": "veridian", |
| "architectures": ["VeridianForCausalLM"], |
| "name_or_path": "MagistrTheOne/veridian-beta", |
| "vocab_size": 128000, |
| "hidden_size": 5120, |
| "intermediate_size": 12288, |
| "num_hidden_layers": 34, |
| "num_attention_heads": 40, |
| "num_key_value_heads": 8, |
| "num_local_experts": 8, |
| "num_shared_experts": 1, |
| "num_experts_per_tok": 2, |
| "max_position_embeddings": 8192, |
| "target_max_position_embeddings": 131072, |
| "pad_token_id": 0, |
| "bos_token_id": 1, |
| "eos_token_id": 2, |
| "attention_dropout": 0.0, |
| "residual_dropout": 0.0, |
| "rms_norm_eps": 1e-06, |
| "rope_theta": 500000.0, |
| "rope_scaling": { |
| "rope_type": "yarn", |
| "factor": 16.0, |
| "original_max_position_embeddings": 8192, |
| "beta_fast": 32.0, |
| "beta_slow": 1.0 |
| }, |
| "router_aux_loss_coef": 0.01, |
| "router_z_loss_coef": 0.0001, |
| "capacity_factor_train": 1.25, |
| "capacity_factor_eval": 1.0, |
| "initializer_range": 0.02, |
| "tie_word_embeddings": true, |
| "use_bias": false, |
| "use_cache": true, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.57.1" |
| } |
|
|