{ "architectures": [ "Llama4ForCausalLM" ], "attention_bias": false, "attention_chunk_size": 32768, "attention_dropout": 0.0, "attn_scale": 0.1, "attn_temperature_tuning": false, "bos_token_id": 128000, "eos_token_id": [ 128001, 128008, 128009 ], "floor_scale": 8192, "for_llm_compressor": false, "head_dim": 64, "hidden_act": "silu", "hidden_size": 1536, "initializer_range": 0.02, "interleave_moe_layer_step": 0, "intermediate_size": 8192, "intermediate_size_mlp": 6144, "layer_types": [ "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention" ], "max_position_embeddings": 32768, "model_type": "llama4_text", "moe_layers": [], "no_rope_layers": [ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 ], "num_attention_heads": 24, "num_experts_per_tok": 0, "num_hidden_layers": 22, "num_key_value_heads": 6, "num_local_experts": 0, "output_router_logits": false, "rms_norm_eps": 1e-05, "rope_scaling": null, "rope_theta": 8000000.0, "router_aux_loss_coef": 0.001, "router_jitter_noise": 0.0, "tie_word_embeddings": true, "torch_dtype": "float32", "transformers_version": "4.55.0", "use_cache": true, "use_qk_norm": true, "vocab_size": 128256, "quantization": { "group_size": 64, "method": "mixed_precision_dynamic", "per_layer_bits": { "layers.0.attention.q_proj": 4, "layers.0.attention.k_proj": 4, "layers.0.attention.v_proj": 4, "layers.0.attention.o_proj": 8, "layers.0.feed_forward.gate_proj": 4, "layers.0.feed_forward.up_proj": 4, "layers.0.feed_forward.down_proj": 4, "layers.1.attention.q_proj": 4, "layers.1.attention.k_proj": 4, "layers.1.attention.v_proj": 4, "layers.1.attention.o_proj": 4, "layers.1.feed_forward.gate_proj": 4, "layers.1.feed_forward.up_proj": 4, "layers.1.feed_forward.down_proj": 4, "layers.2.attention.q_proj": 4, "layers.2.attention.k_proj": 4, "layers.2.attention.v_proj": 4, "layers.2.attention.o_proj": 4, "layers.2.feed_forward.gate_proj": 4, "layers.2.feed_forward.up_proj": 4, "layers.2.feed_forward.down_proj": 4, "layers.3.attention.q_proj": 4, "layers.3.attention.k_proj": 4, "layers.3.attention.v_proj": 4, "layers.3.attention.o_proj": 4, "layers.3.feed_forward.gate_proj": 4, "layers.3.feed_forward.up_proj": 4, "layers.3.feed_forward.down_proj": 4, "layers.4.attention.q_proj": 4, "layers.4.attention.k_proj": 4, "layers.4.attention.v_proj": 4, "layers.4.attention.o_proj": 4, "layers.4.feed_forward.gate_proj": 4, "layers.4.feed_forward.up_proj": 4, "layers.4.feed_forward.down_proj": 4, "layers.5.attention.q_proj": 4, "layers.5.attention.k_proj": 4, "layers.5.attention.v_proj": 4, "layers.5.attention.o_proj": 4, "layers.5.feed_forward.gate_proj": 4, "layers.5.feed_forward.up_proj": 4, "layers.5.feed_forward.down_proj": 4, "layers.6.attention.q_proj": 4, "layers.6.attention.k_proj": 4, "layers.6.attention.v_proj": 4, "layers.6.attention.o_proj": 4, "layers.6.feed_forward.gate_proj": 4, "layers.6.feed_forward.up_proj": 4, "layers.6.feed_forward.down_proj": 4, "layers.7.attention.q_proj": 4, "layers.7.attention.k_proj": 4, "layers.7.attention.v_proj": 4, "layers.7.attention.o_proj": 4, "layers.7.feed_forward.gate_proj": 4, "layers.7.feed_forward.up_proj": 4, "layers.7.feed_forward.down_proj": 4, "layers.8.attention.q_proj": 4, "layers.8.attention.k_proj": 4, "layers.8.attention.v_proj": 4, "layers.8.attention.o_proj": 4, "layers.8.feed_forward.gate_proj": 4, "layers.8.feed_forward.up_proj": 4, "layers.8.feed_forward.down_proj": 4, "layers.9.attention.q_proj": 4, "layers.9.attention.k_proj": 4, "layers.9.attention.v_proj": 4, "layers.9.attention.o_proj": 4, "layers.9.feed_forward.gate_proj": 4, "layers.9.feed_forward.up_proj": 4, "layers.9.feed_forward.down_proj": 4, "layers.10.attention.q_proj": 4, "layers.10.attention.k_proj": 4, "layers.10.attention.v_proj": 4, "layers.10.attention.o_proj": 4, "layers.10.feed_forward.gate_proj": 4, "layers.10.feed_forward.up_proj": 4, "layers.10.feed_forward.down_proj": 4, "layers.11.attention.q_proj": 4, "layers.11.attention.k_proj": 4, "layers.11.attention.v_proj": 4, "layers.11.attention.o_proj": 4, "layers.11.feed_forward.gate_proj": 4, "layers.11.feed_forward.up_proj": 4, "layers.11.feed_forward.down_proj": 4, "layers.12.attention.q_proj": 4, "layers.12.attention.k_proj": 4, "layers.12.attention.v_proj": 4, "layers.12.attention.o_proj": 4, "layers.12.feed_forward.gate_proj": 4, "layers.12.feed_forward.up_proj": 4, "layers.12.feed_forward.down_proj": 4, "layers.13.attention.q_proj": 4, "layers.13.attention.k_proj": 4, "layers.13.attention.v_proj": 4, "layers.13.attention.o_proj": 4, "layers.13.feed_forward.gate_proj": 4, "layers.13.feed_forward.up_proj": 4, "layers.13.feed_forward.down_proj": 4, "layers.14.attention.q_proj": 4, "layers.14.attention.k_proj": 4, "layers.14.attention.v_proj": 4, "layers.14.attention.o_proj": 4, "layers.14.feed_forward.gate_proj": 4, "layers.14.feed_forward.up_proj": 4, "layers.14.feed_forward.down_proj": 4, "layers.15.attention.q_proj": 4, "layers.15.attention.k_proj": 4, "layers.15.attention.v_proj": 4, "layers.15.attention.o_proj": 4, "layers.15.feed_forward.gate_proj": 4, "layers.15.feed_forward.up_proj": 4, "layers.15.feed_forward.down_proj": 4, "layers.16.attention.q_proj": 4, "layers.16.attention.k_proj": 4, "layers.16.attention.v_proj": 4, "layers.16.attention.o_proj": 4, "layers.16.feed_forward.gate_proj": 4, "layers.16.feed_forward.up_proj": 4, "layers.16.feed_forward.down_proj": 4, "layers.17.attention.q_proj": 4, "layers.17.attention.k_proj": 4, "layers.17.attention.v_proj": 4, "layers.17.attention.o_proj": 4, "layers.17.feed_forward.gate_proj": 4, "layers.17.feed_forward.up_proj": 4, "layers.17.feed_forward.down_proj": 4, "layers.18.attention.q_proj": 4, "layers.18.attention.k_proj": 4, "layers.18.attention.v_proj": 4, "layers.18.attention.o_proj": 4, "layers.18.feed_forward.gate_proj": 4, "layers.18.feed_forward.up_proj": 4, "layers.18.feed_forward.down_proj": 4, "layers.19.attention.q_proj": 4, "layers.19.attention.k_proj": 4, "layers.19.attention.v_proj": 4, "layers.19.attention.o_proj": 4, "layers.19.feed_forward.gate_proj": 4, "layers.19.feed_forward.up_proj": 4, "layers.19.feed_forward.down_proj": 4, "layers.20.attention.q_proj": 4, "layers.20.attention.k_proj": 4, "layers.20.attention.v_proj": 4, "layers.20.attention.o_proj": 4, "layers.20.feed_forward.gate_proj": 4, "layers.20.feed_forward.up_proj": 4, "layers.20.feed_forward.down_proj": 4, "layers.21.attention.q_proj": 4, "layers.21.attention.k_proj": 4, "layers.21.attention.v_proj": 4, "layers.21.attention.o_proj": 4, "layers.21.feed_forward.gate_proj": 4, "layers.21.feed_forward.up_proj": 4, "layers.21.feed_forward.down_proj": 4 } } }