name: veridian-1b-bootstrap vocab_size: 32000 pad_token_id: 0 bos_token_id: 1 eos_token_id: 2 max_position_embeddings: 2048 target_max_position_embeddings: 2048 hidden_size: 1536 num_hidden_layers: 16 num_attention_heads: 12 num_key_value_heads: 4 intermediate_size: 4096 num_local_experts: 4 num_shared_experts: 1 num_experts_per_tok: 2 attention_dropout: 0.0 residual_dropout: 0.0 rms_norm_eps: 1.0e-6 rope_theta: 100000.0 router_aux_loss_coef: 0.01 router_z_loss_coef: 1.0e-4 capacity_factor_train: 1.25 capacity_factor_eval: 1.0 initializer_range: 0.02 tie_word_embeddings: true use_bias: false use_cache: true