| { |
| "adapter_reduction_factor": 16, |
| "architectures": [ |
| "ModernALBERTForMaskedLM" |
| ], |
| "attention_probs_dropout_prob": 0.1, |
| "concat_routing_inputs": true, |
| "embedding_size": 128, |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.1, |
| "hidden_size": 1024, |
| "initializer_range": 0.02, |
| "intermediate_size": 2048, |
| "layer_norm_eps": 1e-12, |
| "max_position_embeddings": 8192, |
| "model_type": "ModernALBERT_MoL", |
| "num_attention_heads": 16, |
| "num_expert_layers": 6, |
| "num_hidden_layers": 18, |
| "num_static_groups": 2, |
| "pad_token_id": 0, |
| "router_aux_loss_coef": 0.01, |
| "router_warmup_steps": 7000, |
| "routing_noise_std": 1.0, |
| "routing_noise_type": "gumbel", |
| "static_group_depth": 3, |
| "top_k_experts": 1, |
| "torch_dtype": "float32", |
| "transformers_version": "4.55.2", |
| "use_adapters": true, |
| "use_gated_residual": true, |
| "vocab_size": 50368 |
| } |
|
|
