| { |
| "architectures": [ |
| "Gemma4Eagle3Model" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "attention_k_eq_v": true, |
| "bos_token_id": 2, |
| "draft_num_hidden_layers": 1, |
| "dtype": "bfloat16", |
| "enable_moe_block": false, |
| "eos_token_id": 1, |
| "final_logit_softcapping": 30.0, |
| "global_head_dim": 512, |
| "head_dim": 256, |
| "hidden_activation": "gelu_pytorch_tanh", |
| "hidden_size": 3840, |
| "hidden_size_per_layer_input": 0, |
| "initializer_range": 0.02, |
| "intermediate_size": 15360, |
| "layer_types": [ |
| "full_attention" |
| ], |
| "max_position_embeddings": 262144, |
| "model_type": "gemma4_unified_text", |
| "moe_intermediate_size": null, |
| "num_attention_heads": 16, |
| "num_experts": null, |
| "num_global_key_value_heads": 1, |
| "num_hidden_layers": 1, |
| "num_key_value_heads": 8, |
| "num_kv_shared_layers": 0, |
| "num_target_layers": 48, |
| "pad_token_id": 0, |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "full_attention": { |
| "partial_rotary_factor": 0.25, |
| "rope_theta": 1000000.0, |
| "rope_type": "proportional" |
| }, |
| "rope_theta": null, |
| "rope_type": "default", |
| "sliding_attention": { |
| "rope_theta": 10000.0, |
| "rope_type": "default" |
| } |
| }, |
| "sliding_window": 1024, |
| "step_loss_decay": 0.8, |
| "target_layer_ids": [ |
| 5, |
| 17, |
| 29, |
| 41, |
| 46 |
| ], |
| "target_model_name_or_path": "google/gemma-4-12B-it", |
| "target_model_type": "gemma4_unified", |
| "target_text_model_type": "gemma4_unified_text", |
| "tie_word_embeddings": false, |
| "top_k_experts": null, |
| "transformers_version": "5.10.2", |
| "ttt_length": 7, |
| "use_bidirectional_attention": "vision", |
| "use_cache": true, |
| "use_double_wide_mlp": false, |
| "vocab_size": 262144, |
| "vocab_size_per_layer_input": 262144 |
| } |
|
|
