diff --git "a/model.safetensors.index.json" "b/model.safetensors.index.json" new file mode 100644--- /dev/null +++ "b/model.safetensors.index.json" @@ -0,0 +1,1866 @@ +{ + "metadata": { + "total_size": 13485240320 + }, + "weight_map": { + "lm_head.weight": "model-00003-of-00003.safetensors", + "model.embed_tokens.weight": "model-00001-of-00003.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.0.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.0.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.1.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.1.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.10.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.10.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.11.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.11.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.12.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.12.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.13.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.13.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.14.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.14.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.15.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.15.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.16.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.16.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.17.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.17.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.18.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.18.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.19.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.19.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.2.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.2.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.20.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.20.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.21.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.21.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.22.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.22.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.0": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.1": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.10": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.11": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.12": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.13": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.14": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.15": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.2": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.3": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.4": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.5": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.6": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.7": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.8": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_down.9": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.0": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.1": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.10": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.11": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.12": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.13": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.14": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.15": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.2": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.3": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.4": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.5": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.6": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.7": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.8": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_gate.9": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.0": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.1": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.10": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.11": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.12": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.13": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.14": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.15": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.2": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.3": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.4": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.5": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.6": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.7": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.8": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.calculator.experts.weight_up.9": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.gate.gate_network.0.weight": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.gate.gate_network.2.weight": "model-00002-of-00003.safetensors", + "model.layers.23.mlp.gate.weight_noise.weight": "model-00002-of-00003.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00003.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.23.self_attn.rotary_emb.inv_freq": "model-00002-of-00003.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.0": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.1": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.10": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.11": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.12": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.13": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.14": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.15": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.2": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.3": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.4": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.5": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.6": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.7": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.8": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_down.9": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.0": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.1": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.10": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.11": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.12": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.13": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.14": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.15": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.2": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.3": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.4": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.5": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.6": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.7": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.8": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_gate.9": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.0": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.1": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.10": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.11": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.12": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.13": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.14": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.15": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.2": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.3": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.4": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.5": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.6": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.7": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.8": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.calculator.experts.weight_up.9": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.gate.gate_network.0.weight": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.gate.gate_network.2.weight": "model-00003-of-00003.safetensors", + "model.layers.24.mlp.gate.weight_noise.weight": "model-00003-of-00003.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.24.self_attn.rotary_emb.inv_freq": "model-00003-of-00003.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.0": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.1": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.10": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.11": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.12": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.13": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.14": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.15": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.2": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.3": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.4": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.5": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.6": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.7": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.8": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_down.9": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.0": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.1": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.10": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.11": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.12": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.13": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.14": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.15": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.2": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.3": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.4": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.5": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.6": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.7": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.8": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_gate.9": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.0": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.1": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.10": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.11": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.12": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.13": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.14": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.15": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.2": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.3": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.4": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.5": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.6": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.7": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.8": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.calculator.experts.weight_up.9": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.gate.gate_network.0.weight": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.gate.gate_network.2.weight": "model-00003-of-00003.safetensors", + "model.layers.25.mlp.gate.weight_noise.weight": "model-00003-of-00003.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.25.self_attn.rotary_emb.inv_freq": "model-00003-of-00003.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.0": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.1": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.10": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.11": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.12": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.13": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.14": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.15": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.2": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.3": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.4": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.5": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.6": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.7": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.8": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_down.9": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.0": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.1": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.10": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.11": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.12": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.13": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.14": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.15": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.2": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.3": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.4": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.5": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.6": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.7": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.8": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_gate.9": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.0": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.1": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.10": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.11": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.12": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.13": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.14": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.15": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.2": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.3": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.4": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.5": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.6": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.7": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.8": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.calculator.experts.weight_up.9": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.gate.gate_network.0.weight": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.gate.gate_network.2.weight": "model-00003-of-00003.safetensors", + "model.layers.26.mlp.gate.weight_noise.weight": "model-00003-of-00003.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.26.self_attn.rotary_emb.inv_freq": "model-00003-of-00003.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.0": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.1": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.10": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.11": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.12": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.13": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.14": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.15": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.2": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.3": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.4": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.5": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.6": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.7": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.8": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_down.9": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.0": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.1": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.10": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.11": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.12": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.13": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.14": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.15": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.2": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.3": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.4": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.5": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.6": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.7": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.8": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_gate.9": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.0": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.1": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.10": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.11": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.12": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.13": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.14": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.15": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.2": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.3": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.4": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.5": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.6": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.7": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.8": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.calculator.experts.weight_up.9": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.gate.gate_network.0.weight": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.gate.gate_network.2.weight": "model-00003-of-00003.safetensors", + "model.layers.27.mlp.gate.weight_noise.weight": "model-00003-of-00003.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.27.self_attn.rotary_emb.inv_freq": "model-00003-of-00003.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.0": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.1": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.10": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.11": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.12": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.13": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.14": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.15": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.2": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.3": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.4": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.5": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.6": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.7": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.8": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_down.9": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.0": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.1": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.10": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.11": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.12": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.13": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.14": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.15": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.2": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.3": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.4": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.5": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.6": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.7": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.8": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_gate.9": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.0": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.1": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.10": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.11": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.12": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.13": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.14": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.15": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.2": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.3": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.4": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.5": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.6": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.7": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.8": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.calculator.experts.weight_up.9": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.gate.gate_network.0.weight": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.gate.gate_network.2.weight": "model-00003-of-00003.safetensors", + "model.layers.28.mlp.gate.weight_noise.weight": "model-00003-of-00003.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.28.self_attn.rotary_emb.inv_freq": "model-00003-of-00003.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.0": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.1": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.10": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.11": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.12": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.13": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.14": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.15": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.2": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.3": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.4": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.5": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.6": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.7": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.8": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_down.9": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.0": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.1": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.10": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.11": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.12": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.13": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.14": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.15": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.2": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.3": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.4": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.5": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.6": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.7": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.8": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_gate.9": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.0": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.1": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.10": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.11": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.12": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.13": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.14": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.15": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.2": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.3": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.4": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.5": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.6": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.7": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.8": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.calculator.experts.weight_up.9": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.gate.gate_network.0.weight": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.gate.gate_network.2.weight": "model-00003-of-00003.safetensors", + "model.layers.29.mlp.gate.weight_noise.weight": "model-00003-of-00003.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.29.self_attn.rotary_emb.inv_freq": "model-00003-of-00003.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.3.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.3.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.0": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.1": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.10": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.11": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.12": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.13": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.14": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.15": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.2": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.3": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.4": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.5": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.6": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.7": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.8": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_down.9": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.0": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.1": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.10": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.11": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.12": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.13": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.14": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.15": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.2": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.3": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.4": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.5": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.6": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.7": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.8": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_gate.9": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.0": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.1": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.10": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.11": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.12": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.13": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.14": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.15": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.2": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.3": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.4": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.5": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.6": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.7": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.8": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.calculator.experts.weight_up.9": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.gate.gate_network.0.weight": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.gate.gate_network.2.weight": "model-00003-of-00003.safetensors", + "model.layers.30.mlp.gate.weight_noise.weight": "model-00003-of-00003.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.30.self_attn.rotary_emb.inv_freq": "model-00003-of-00003.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.0": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.1": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.10": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.11": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.12": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.13": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.14": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.15": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.2": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.3": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.4": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.5": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.6": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.7": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.8": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_down.9": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.0": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.1": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.10": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.11": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.12": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.13": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.14": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.15": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.2": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.3": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.4": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.5": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.6": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.7": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.8": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_gate.9": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.0": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.1": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.10": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.11": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.12": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.13": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.14": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.15": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.2": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.3": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.4": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.5": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.6": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.7": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.8": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.calculator.experts.weight_up.9": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.gate.gate_network.0.weight": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.gate.gate_network.2.weight": "model-00003-of-00003.safetensors", + "model.layers.31.mlp.gate.weight_noise.weight": "model-00003-of-00003.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.31.self_attn.rotary_emb.inv_freq": "model-00003-of-00003.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.4.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.4.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.5.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.5.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.6.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.6.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.7.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.7.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.8.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.8.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.0": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.1": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.10": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.11": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.12": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.13": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.14": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.15": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.2": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.3": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.4": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.5": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.6": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.7": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.8": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_down.9": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.0": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.1": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.10": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.11": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.12": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.13": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.14": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.15": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.2": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.3": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.4": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.5": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.6": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.7": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.8": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_gate.9": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.0": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.1": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.10": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.11": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.12": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.13": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.14": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.15": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.2": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.3": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.4": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.5": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.6": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.7": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.8": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.calculator.experts.weight_up.9": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.gate.gate_network.0.weight": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.gate.gate_network.2.weight": "model-00001-of-00003.safetensors", + "model.layers.9.mlp.gate.weight_noise.weight": "model-00001-of-00003.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors", + "model.layers.9.self_attn.rotary_emb.inv_freq": "model-00001-of-00003.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors", + "model.norm.weight": "model-00003-of-00003.safetensors" + } +}