| { |
| "architectures": [ |
| "MoedlForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "auto_map": { |
| "AutoConfig": "configuration_moedl.MoedlConfig", |
| "AutoModelForCausalLM": "modeling_moedl.MoedlForCausalLM" |
| }, |
| "bos_token_id": 1, |
| "capacity_factor": 0.0, |
| "dtype": "float32", |
| "eos_token_id": 2, |
| "head_dim": 128, |
| "hidden_act": "silu", |
| "hidden_size": 768, |
| "initializer_range": 0.02, |
| "intermediate_size": 2048, |
| "lb_coeff": 0.0, |
| "lb_gamma": 0.0, |
| "max_position_embeddings": 512, |
| "mlp_bias": false, |
| "model_type": "moedl", |
| "n_params": "137M", |
| "num_active_experts": 1, |
| "num_attention_heads": 16, |
| "num_experts": 1, |
| "num_hidden_layers": 8, |
| "num_key_value_heads": 16, |
| "num_shared_experts": 0, |
| "pad_token_id": 2, |
| "pretraining_tp": 1, |
| "rms_norm_eps": 1e-06, |
| "rope_scaling": null, |
| "rope_theta": 10000.0, |
| "tie_word_embeddings": false, |
| "transformers_version": "4.57.3", |
| "use_cache": true, |
| "vocab_size": 32000 |
| } |
|
|
