Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+model-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,3 @@
+# GLM-4.7-Flash-small-test
+
+For testing only

chat_template.jinja

@@ -0,0 +1,86 @@
+[gMASK]<sop>
+{%- if tools -%}
+<|system|>
+# Tools
+
+You may call one or more functions to assist with the user query.
+
+You are provided with function signatures within <tools></tools> XML tags:
+<tools>
+{% for tool in tools %}
+{{ tool | tojson(ensure_ascii=False) }}
+{% endfor %}
+</tools>
+
+For each function call, output the function name and arguments within the following XML format:
+<tool_call>{function-name}<arg_key>{arg-key-1}</arg_key><arg_value>{arg-value-1}</arg_value><arg_key>{arg-key-2}</arg_key><arg_value>{arg-value-2}</arg_value>...</tool_call>{%- endif -%}
+{%- macro visible_text(content) -%}
+    {%- if content is string -%}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping -%}
+        {%- for item in content -%}
+            {%- if item is mapping and item.type == 'text' -%}
+                {{- item.text }}
+            {%- elif item is string -%}
+                {{- item }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{- content }}
+    {%- endif -%}
+{%- endmacro -%}
+{%- set ns = namespace(last_user_index=-1) %}
+{%- for m in messages %}
+    {%- if m.role == 'user' %}
+        {% set ns.last_user_index = loop.index0 -%}
+    {%- endif %}
+{%- endfor %}
+{% for m in messages %}
+{%- if m.role == 'user' -%}<|user|>{{ visible_text(m.content) }}
+{%- elif m.role == 'assistant' -%}
+<|assistant|>
+{%- set reasoning_content = '' %}
+{%- set content = visible_text(m.content) %}
+{%- if m.reasoning_content is string %}
+    {%- set reasoning_content = m.reasoning_content %}
+{%- else %}
+    {%- if '</think>' in content %}
+        {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+        {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+    {%- endif %}
+{%- endif %}
+{%- if ((clear_thinking is defined and not clear_thinking) or loop.index0 > ns.last_user_index) and reasoning_content -%}
+{{ '<think>' + reasoning_content.strip() +  '</think>'}}
+{%- else -%}
+{{ '</think>' }}
+{%- endif -%}
+{%- if content.strip() -%}
+{{ content.strip() }}
+{%- endif -%}
+{% if m.tool_calls %}
+{% for tc in m.tool_calls %}
+{%- if tc.function %}
+    {%- set tc = tc.function %}
+{%- endif %}
+{{- '<tool_call>' + tc.name -}}
+{% set _args = tc.arguments %}{% for k, v in _args.items() %}<arg_key>{{ k }}</arg_key><arg_value>{{ v | tojson(ensure_ascii=False) if v is not string else v }}</arg_value>{% endfor %}</tool_call>{% endfor %}
+{% endif %}
+{%- elif m.role == 'tool' -%}
+{%- if m.content is string -%}
+{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+    {{- '<|observation|>' }}
+{%- endif %}
+{{- '<tool_response>' }}
+{{- m.content }}
+{{- '</tool_response>' }}
+{%- else -%}
+<|observation|>{% for tr in m.content %}
+<tool_response>{{ tr.output if tr.output is defined else tr }}</tool_response>{% endfor -%}
+{% endif -%}
+{%- elif m.role == 'system' -%}
+<|system|>{{ visible_text(m.content) }}
+{%- endif -%}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    <|assistant|>{{- '</think>' if (enable_thinking is defined and not enable_thinking) else '<think>' -}}
+{%- endif -%}

config.json

@@ -0,0 +1,45 @@
+{
+  "architectures": [
+    "Glm4MoeLiteForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "pad_token_id": 154820,
+  "eos_token_id": [
+    154820,
+    154827,
+    154829
+  ],
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "intermediate_size": 10240,
+  "max_position_embeddings": 202752,
+  "model_type": "glm4_moe_lite",
+  "moe_intermediate_size": 1536,
+  "topk_method": "noaux_tc",
+  "norm_topk_prob": true,
+  "num_attention_heads": 20,
+  "n_group": 1,
+  "topk_group": 1,
+  "n_routed_experts": 2,
+  "n_shared_experts": 1,
+  "routed_scaling_factor": 1.8,
+  "num_experts_per_tok": 2,
+  "first_k_dense_replace": 1,
+  "num_hidden_layers": 47,
+  "num_key_value_heads": 20,
+  "num_nextn_predict_layers": 1,
+  "partial_rotary_factor": 1.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 1000000,
+  "tie_word_embeddings": false,
+  "dtype": "bfloat16",
+  "transformers_version": "5.0.0rc0",
+  "q_lora_rank": 768,
+  "kv_lora_rank": 512,
+  "qk_nope_head_dim": 192,
+  "qk_rope_head_dim": 64,
+  "v_head_dim": 256,
+  "vocab_size": 154880
+}

generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": [
+    154820,
+    154827,
+    154829
+  ],
+  "pad_token_id": 154820,
+  "temperature": 1.0,
+  "transformers_version": "5.0.0.dev0"
+}

make.py

@@ -0,0 +1,109 @@
+import os
+import json
+import re
+from pathlib import Path
+from safetensors import safe_open
+from safetensors.torch import save_file
+import torch
+
+def main():
+  src_dir = Path("../GLM-4.7-Flash")
+  dst_path = Path("model.safetensors")
+  num_experts_to_keep = 2
+  
+  # Find all safetensors files
+  safetensor_files = sorted(src_dir.glob("*.safetensors"))
+  print(f"Found {len(safetensor_files)} safetensors files")
+  
+  # Pattern to match expert weights
+  expert_pattern = re.compile(r"model\.layers\.(\d+)\.mlp\.experts\.(\d+)\..+")
+  gate_pattern = re.compile(r"model\.layers\.(\d+)\.mlp\.gate\.weight")
+  bias_pattern = re.compile(r"model\.layers\.(\d+)\.mlp\.gate\.e_score_correction_bias")
+  
+  new_tensors = {}
+  
+  for sf_path in safetensor_files:
+    print(f"Processing {sf_path.name}...")
+    
+    with safe_open(sf_path, framework="pt", device="cpu") as f:
+      for key in f.keys():
+        tensor = f.get_tensor(key)
+        
+        # Check if this is an expert weight
+        expert_match = expert_pattern.search(key)
+        if expert_match:
+          layer_idx = int(expert_match.group(1))
+          expert_idx = int(expert_match.group(2))
+          
+          if expert_idx >= num_experts_to_keep:
+            print(f"  Skipping {key} (expert {expert_idx} >= {num_experts_to_keep})")
+            continue
+          
+          new_tensors[key] = tensor
+          continue
+        
+        # Check if this is the gate weight
+        gate_match = gate_pattern.search(key)
+        if gate_match:
+          layer_idx = int(gate_match.group(1))
+          original_shape = tensor.shape
+          # Gate weight is [num_experts, hidden_size], keep first 8 experts
+          new_tensor = tensor[:num_experts_to_keep, :]
+          print(f"  Resizing {key}: {original_shape} -> {new_tensor.shape}")
+          new_tensors[key] = new_tensor
+          continue
+        
+        # Check if this is the e_score_correction_bias
+        bias_match = bias_pattern.search(key)
+        if bias_match:
+          layer_idx = int(bias_match.group(1))
+          original_shape = tensor.shape
+          # Bias is [num_experts], keep first 8
+          new_tensor = tensor[:num_experts_to_keep]
+          print(f"  Resizing {key}: {original_shape} -> {new_tensor.shape}")
+          new_tensors[key] = new_tensor
+          continue
+        
+        # Keep all other tensors as-is
+        new_tensors[key] = tensor
+  
+  print(f"\nTotal tensors to save: {len(new_tensors)}")
+  print(f"Saving to {dst_path}...")
+  
+  save_file(new_tensors, dst_path)
+  print("Done!")
+  
+  # Also copy and modify config.json
+  config_src = src_dir / "config.json"
+  if config_src.exists():
+    with open(config_src, "r") as f:
+      config = json.load(f)
+    
+    # Update number of experts
+    # if "num_experts" in config:
+    #   print(f"Updating num_experts: {config['num_experts']} -> {num_experts_to_keep}")
+    #   config["num_experts"] = num_experts_to_keep
+    # if "n_routed_experts" in config:
+    #   print(f"Updating n_routed_experts: {config['n_routed_experts']} -> {num_experts_to_keep}")
+    #   config["n_routed_experts"] = num_experts_to_keep
+    
+    # config_dst = Path("config.json")
+    # with open(config_dst, "w") as f:
+    #   json.dump(config, f, indent=2)
+    # print(f"Saved modified config to {config_dst}")
+  
+  # Create safetensors index file
+  index_data = {
+    "metadata": {
+      "total_size": sum(t.numel() * t.element_size() for t in new_tensors.values())
+    },
+    "weight_map": {key: str(dst_path) for key in new_tensors.keys()}
+  }
+  
+  index_dst = Path("model.safetensors.index.json")
+  with open(index_dst, "w") as f:
+    json.dump(index_data, f, indent=2)
+  print(f"Saved safetensors index to {index_dst}")
+
+if __name__ == "__main__":
+  main()

model-Q8_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5087e640fc8c9411aaeaf83e3f6dd13fe36d80a55b5444554799a3b70fe6b2a6
+size 3222398304

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a0c02a05a0be2138810e34771a95148c633a41d2344694fdc940dc5b9bec86a
+size 7431243408

model.safetensors.index.json

@@ -0,0 +1,968 @@
+{
+  "metadata": {
+    "total_size": 7431127416
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model.safetensors",
+    "model.layers.0.input_layernorm.weight": "model.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.0.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.0.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.0.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.0.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.0.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.0.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.47.embed_tokens.weight": "model.safetensors",
+    "model.layers.1.input_layernorm.weight": "model.safetensors",
+    "model.layers.1.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.1.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.1.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.1.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.1.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.1.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.1.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.1.mlp.gate.weight": "model.safetensors",
+    "model.layers.1.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.1.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.1.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.1.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.1.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.1.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.1.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.1.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.1.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.2.input_layernorm.weight": "model.safetensors",
+    "model.layers.2.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.2.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.2.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.2.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.2.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.2.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.2.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.2.mlp.gate.weight": "model.safetensors",
+    "model.layers.2.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.2.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.2.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.2.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.2.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.2.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.2.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.2.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.2.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.3.input_layernorm.weight": "model.safetensors",
+    "model.layers.3.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.3.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.3.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.3.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.3.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.3.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.3.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.3.mlp.gate.weight": "model.safetensors",
+    "model.layers.3.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.3.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.3.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.3.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.3.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.3.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.3.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.3.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.3.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.4.input_layernorm.weight": "model.safetensors",
+    "model.layers.4.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.4.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.4.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.4.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.4.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.4.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.4.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.4.mlp.gate.weight": "model.safetensors",
+    "model.layers.4.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.4.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.4.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.4.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.4.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.4.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.4.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.4.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.4.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.5.input_layernorm.weight": "model.safetensors",
+    "model.layers.5.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.5.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.5.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.5.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.5.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.5.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.5.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.5.mlp.gate.weight": "model.safetensors",
+    "model.layers.5.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.5.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.5.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.5.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.5.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.5.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.5.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.5.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.5.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.6.input_layernorm.weight": "model.safetensors",
+    "model.layers.6.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.6.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.6.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.6.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.6.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.6.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.6.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.6.mlp.gate.weight": "model.safetensors",
+    "model.layers.6.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.6.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.6.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.6.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.6.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.6.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.6.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.6.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.6.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.7.input_layernorm.weight": "model.safetensors",
+    "model.layers.7.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.7.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.7.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.7.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.7.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.7.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.7.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.7.mlp.gate.weight": "model.safetensors",
+    "model.layers.7.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.7.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.7.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.7.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.7.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.7.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.7.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.7.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.7.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.8.input_layernorm.weight": "model.safetensors",
+    "model.layers.8.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.8.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.8.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.8.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.8.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.8.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.8.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.8.mlp.gate.weight": "model.safetensors",
+    "model.layers.8.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.8.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.8.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.8.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.8.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.8.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.8.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.8.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.8.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.9.input_layernorm.weight": "model.safetensors",
+    "model.layers.9.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.9.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.9.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.9.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.9.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.9.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.9.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.9.mlp.gate.weight": "model.safetensors",
+    "model.layers.9.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.9.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.9.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.9.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.9.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.9.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.9.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.9.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.9.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.10.input_layernorm.weight": "model.safetensors",
+    "model.layers.10.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.10.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.10.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.10.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.10.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.10.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.10.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.10.mlp.gate.weight": "model.safetensors",
+    "model.layers.10.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.10.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.10.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.10.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.10.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.10.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.10.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.10.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.10.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.11.input_layernorm.weight": "model.safetensors",
+    "model.layers.11.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.11.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.11.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.11.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.11.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.11.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.11.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.11.mlp.gate.weight": "model.safetensors",
+    "model.layers.11.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.11.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.11.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.11.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.11.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.11.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.11.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.11.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.11.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.12.input_layernorm.weight": "model.safetensors",
+    "model.layers.12.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.12.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.12.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.12.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.12.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.12.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.12.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.12.mlp.gate.weight": "model.safetensors",
+    "model.layers.12.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.12.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.12.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.12.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.12.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.12.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.12.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.12.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.12.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.13.input_layernorm.weight": "model.safetensors",
+    "model.layers.13.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.13.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.13.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.13.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.13.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.13.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.13.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.13.mlp.gate.weight": "model.safetensors",
+    "model.layers.13.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.13.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.13.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.13.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.13.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.13.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.13.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.13.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.13.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.14.input_layernorm.weight": "model.safetensors",
+    "model.layers.14.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.14.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.14.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.14.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.14.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.14.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.14.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.14.mlp.gate.weight": "model.safetensors",
+    "model.layers.14.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.14.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.14.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.14.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.14.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.14.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.14.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.14.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.14.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.15.input_layernorm.weight": "model.safetensors",
+    "model.layers.15.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.15.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.15.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.15.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.15.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.15.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.15.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.15.mlp.gate.weight": "model.safetensors",
+    "model.layers.15.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.15.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.15.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.15.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.15.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.15.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.15.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.15.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.15.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.16.input_layernorm.weight": "model.safetensors",
+    "model.layers.16.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.16.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.16.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.16.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.16.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.16.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.16.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.16.mlp.gate.weight": "model.safetensors",
+    "model.layers.16.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.16.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.16.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.16.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.16.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.16.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.16.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.16.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.16.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.17.input_layernorm.weight": "model.safetensors",
+    "model.layers.17.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.17.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.17.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.17.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.17.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.17.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.17.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.17.mlp.gate.weight": "model.safetensors",
+    "model.layers.17.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.17.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.17.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.17.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.17.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.17.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.17.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.17.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.17.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.18.input_layernorm.weight": "model.safetensors",
+    "model.layers.18.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.18.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.18.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.18.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.18.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.18.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.18.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.18.mlp.gate.weight": "model.safetensors",
+    "model.layers.18.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.18.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.18.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.18.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.18.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.18.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.18.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.18.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.18.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.19.input_layernorm.weight": "model.safetensors",
+    "model.layers.19.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.19.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.19.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.19.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.19.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.19.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.19.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.19.mlp.gate.weight": "model.safetensors",
+    "model.layers.19.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.19.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.19.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.19.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.19.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.19.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.19.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.19.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.19.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.20.input_layernorm.weight": "model.safetensors",
+    "model.layers.20.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.20.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.20.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.20.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.20.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.20.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.20.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.20.mlp.gate.weight": "model.safetensors",
+    "model.layers.20.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.20.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.20.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.20.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.20.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.20.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.20.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.20.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.20.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.21.input_layernorm.weight": "model.safetensors",
+    "model.layers.21.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.21.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.21.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.21.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.21.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.21.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.21.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.21.mlp.gate.weight": "model.safetensors",
+    "model.layers.21.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.21.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.21.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.21.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.21.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.21.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.21.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.21.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.21.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.22.input_layernorm.weight": "model.safetensors",
+    "model.layers.22.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.22.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.22.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.22.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.22.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.22.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.22.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.22.mlp.gate.weight": "model.safetensors",
+    "model.layers.22.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.22.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.22.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.22.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.22.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.22.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.22.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.22.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.22.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.23.input_layernorm.weight": "model.safetensors",
+    "model.layers.23.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.23.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.23.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.23.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.23.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.23.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.23.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.23.mlp.gate.weight": "model.safetensors",
+    "model.layers.23.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.23.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.23.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.23.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.23.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.23.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.23.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.23.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.23.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.24.input_layernorm.weight": "model.safetensors",
+    "model.layers.24.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.24.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.24.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.24.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.24.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.24.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.24.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.24.mlp.gate.weight": "model.safetensors",
+    "model.layers.24.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.24.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.24.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.24.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.24.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.24.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.24.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.24.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.24.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.25.input_layernorm.weight": "model.safetensors",
+    "model.layers.25.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.25.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.25.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.25.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.25.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.25.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.25.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.25.mlp.gate.weight": "model.safetensors",
+    "model.layers.25.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.25.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.25.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.25.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.25.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.25.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.25.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.25.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.25.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.26.input_layernorm.weight": "model.safetensors",
+    "model.layers.26.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.26.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.26.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.26.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.26.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.26.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.26.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.26.mlp.gate.weight": "model.safetensors",
+    "model.layers.26.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.26.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.26.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.26.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.26.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.26.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.26.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.26.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.26.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.27.input_layernorm.weight": "model.safetensors",
+    "model.layers.27.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.27.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.27.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.27.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.27.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.27.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.27.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.27.mlp.gate.weight": "model.safetensors",
+    "model.layers.27.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.27.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.27.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.27.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.27.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.27.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.27.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.27.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.27.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.28.input_layernorm.weight": "model.safetensors",
+    "model.layers.28.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.28.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.28.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.28.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.28.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.28.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.28.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.28.mlp.gate.weight": "model.safetensors",
+    "model.layers.28.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.28.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.28.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.28.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.28.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.28.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.28.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.28.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.28.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.29.input_layernorm.weight": "model.safetensors",
+    "model.layers.29.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.29.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.29.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.29.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.29.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.29.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.29.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.29.mlp.gate.weight": "model.safetensors",
+    "model.layers.29.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.29.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.29.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.29.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.29.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.29.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.29.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.29.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.29.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.30.input_layernorm.weight": "model.safetensors",
+    "model.layers.30.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.30.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.30.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.30.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.30.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.30.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.30.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.30.mlp.gate.weight": "model.safetensors",
+    "model.layers.30.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.30.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.30.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.30.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.30.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.30.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.30.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.30.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.30.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.31.input_layernorm.weight": "model.safetensors",
+    "model.layers.31.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.31.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.31.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.31.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.31.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.31.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.31.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.31.mlp.gate.weight": "model.safetensors",
+    "model.layers.31.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.31.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.31.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.31.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.31.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.31.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.31.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.31.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.31.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.32.input_layernorm.weight": "model.safetensors",
+    "model.layers.32.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.32.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.32.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.32.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.32.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.32.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.32.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.32.mlp.gate.weight": "model.safetensors",
+    "model.layers.32.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.32.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.32.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.32.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.32.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.32.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.32.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.32.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.32.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.33.input_layernorm.weight": "model.safetensors",
+    "model.layers.33.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.33.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.33.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.33.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.33.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.33.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.33.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.33.mlp.gate.weight": "model.safetensors",
+    "model.layers.33.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.33.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.33.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.33.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.33.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.33.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.33.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.33.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.33.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.34.input_layernorm.weight": "model.safetensors",
+    "model.layers.34.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.34.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.34.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.34.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.34.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.34.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.34.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.34.mlp.gate.weight": "model.safetensors",
+    "model.layers.34.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.34.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.34.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.34.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.34.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.34.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.34.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.34.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.34.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.35.input_layernorm.weight": "model.safetensors",
+    "model.layers.35.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.35.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.35.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.35.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.35.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.35.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.35.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.35.mlp.gate.weight": "model.safetensors",
+    "model.layers.35.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.35.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.35.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.35.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.35.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.35.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.35.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.35.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.35.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.36.input_layernorm.weight": "model.safetensors",
+    "model.layers.36.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.36.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.36.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.36.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.36.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.36.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.36.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.36.mlp.gate.weight": "model.safetensors",
+    "model.layers.36.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.36.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.36.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.36.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.36.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.36.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.36.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.36.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.36.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.37.input_layernorm.weight": "model.safetensors",
+    "model.layers.37.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.37.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.37.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.37.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.37.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.37.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.37.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.37.mlp.gate.weight": "model.safetensors",
+    "model.layers.37.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.37.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.37.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.37.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.37.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.37.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.37.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.37.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.37.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.38.input_layernorm.weight": "model.safetensors",
+    "model.layers.38.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.38.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.38.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.38.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.38.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.38.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.38.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.38.mlp.gate.weight": "model.safetensors",
+    "model.layers.38.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.38.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.38.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.38.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.38.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.38.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.38.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.38.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.38.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.39.input_layernorm.weight": "model.safetensors",
+    "model.layers.39.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.39.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.39.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.39.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.39.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.39.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.39.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.39.mlp.gate.weight": "model.safetensors",
+    "model.layers.39.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.39.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.39.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.39.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.39.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.39.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.39.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.39.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.39.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.40.input_layernorm.weight": "model.safetensors",
+    "model.layers.40.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.40.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.40.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.40.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.40.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.40.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.40.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.40.mlp.gate.weight": "model.safetensors",
+    "model.layers.40.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.40.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.40.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.40.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.40.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.40.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.40.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.40.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.40.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.41.input_layernorm.weight": "model.safetensors",
+    "model.layers.41.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.41.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.41.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.41.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.41.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.41.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.41.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.41.mlp.gate.weight": "model.safetensors",
+    "model.layers.41.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.41.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.41.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.41.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.41.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.41.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.41.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.41.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.41.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.42.input_layernorm.weight": "model.safetensors",
+    "model.layers.42.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.42.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.42.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.42.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.42.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.42.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.42.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.42.mlp.gate.weight": "model.safetensors",
+    "model.layers.42.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.42.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.42.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.42.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.42.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.42.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.42.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.42.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.42.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.43.input_layernorm.weight": "model.safetensors",
+    "model.layers.43.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.43.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.43.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.43.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.43.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.43.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.43.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.43.mlp.gate.weight": "model.safetensors",
+    "model.layers.43.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.43.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.43.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.43.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.43.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.43.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.43.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.43.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.43.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.44.input_layernorm.weight": "model.safetensors",
+    "model.layers.44.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.44.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.44.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.44.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.44.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.44.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.44.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.44.mlp.gate.weight": "model.safetensors",
+    "model.layers.44.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.44.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.44.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.44.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.44.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.44.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.44.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.44.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.44.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.45.input_layernorm.weight": "model.safetensors",
+    "model.layers.45.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.45.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.45.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.45.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.45.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.45.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.45.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.45.mlp.gate.weight": "model.safetensors",
+    "model.layers.45.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.45.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.45.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.45.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.45.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.45.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.45.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.45.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.45.self_attn.q_b_proj.weight": "model.safetensors",
+    "lm_head.weight": "model.safetensors",
+    "model.layers.46.input_layernorm.weight": "model.safetensors",
+    "model.layers.46.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.46.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.46.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.46.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.46.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.46.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.46.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.46.mlp.gate.weight": "model.safetensors",
+    "model.layers.46.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.46.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.46.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.46.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.46.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.46.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.46.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.46.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.46.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.46.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.46.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.47.shared_head.head.weight": "model.safetensors",
+    "model.norm.weight": "model.safetensors",
+    "model.layers.47.eh_proj.weight": "model.safetensors",
+    "model.layers.47.enorm.weight": "model.safetensors",
+    "model.layers.47.hnorm.weight": "model.safetensors",
+    "model.layers.47.input_layernorm.weight": "model.safetensors",
+    "model.layers.47.mlp.experts.0.down_proj.weight": "model.safetensors",
+    "model.layers.47.mlp.experts.0.gate_proj.weight": "model.safetensors",
+    "model.layers.47.mlp.experts.0.up_proj.weight": "model.safetensors",
+    "model.layers.47.mlp.experts.1.down_proj.weight": "model.safetensors",
+    "model.layers.47.mlp.experts.1.gate_proj.weight": "model.safetensors",
+    "model.layers.47.mlp.experts.1.up_proj.weight": "model.safetensors",
+    "model.layers.47.mlp.gate.e_score_correction_bias": "model.safetensors",
+    "model.layers.47.mlp.gate.weight": "model.safetensors",
+    "model.layers.47.mlp.shared_experts.down_proj.weight": "model.safetensors",
+    "model.layers.47.mlp.shared_experts.gate_proj.weight": "model.safetensors",
+    "model.layers.47.mlp.shared_experts.up_proj.weight": "model.safetensors",
+    "model.layers.47.post_attention_layernorm.weight": "model.safetensors",
+    "model.layers.47.self_attn.kv_a_layernorm.weight": "model.safetensors",
+    "model.layers.47.self_attn.kv_a_proj_with_mqa.weight": "model.safetensors",
+    "model.layers.47.self_attn.kv_b_proj.weight": "model.safetensors",
+    "model.layers.47.self_attn.o_proj.weight": "model.safetensors",
+    "model.layers.47.self_attn.q_a_layernorm.weight": "model.safetensors",
+    "model.layers.47.self_attn.q_a_proj.weight": "model.safetensors",
+    "model.layers.47.self_attn.q_b_proj.weight": "model.safetensors",
+    "model.layers.47.shared_head.norm.weight": "model.safetensors"
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19e773648cb4e65de8660ea6365e10acca112d42a854923df93db4a6f333a82d
+size 20217442

tokenizer_config.json

@@ -0,0 +1,321 @@
+{
+  "added_tokens_decoder": {
+    "154820": {
+      "content": "<|endoftext|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154821": {
+      "content": "[MASK]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154822": {
+      "content": "[gMASK]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154823": {
+      "content": "[sMASK]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154824": {
+      "content": "<sop>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154825": {
+      "content": "<eop>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154826": {
+      "content": "<|system|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154827": {
+      "content": "<|user|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154828": {
+      "content": "<|assistant|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154829": {
+      "content": "<|observation|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154830": {
+      "content": "<|begin_of_image|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154831": {
+      "content": "<|end_of_image|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154832": {
+      "content": "<|begin_of_video|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154833": {
+      "content": "<|end_of_video|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154834": {
+      "content": "<|begin_of_audio|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154835": {
+      "content": "<|end_of_audio|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154836": {
+      "content": "<|begin_of_transcription|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154837": {
+      "content": "<|end_of_transcription|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "154838": {
+      "content": "<|code_prefix|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154839": {
+      "content": "<|code_middle|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154840": {
+      "content": "<|code_suffix|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154841": {
+      "content": "<think>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154842": {
+      "content": "</think>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154843": {
+      "content": "<tool_call>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154844": {
+      "content": "</tool_call>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154845": {
+      "content": "<tool_response>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154846": {
+      "content": "</tool_response>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154847": {
+      "content": "<arg_key>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154848": {
+      "content": "</arg_key>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154849": {
+      "content": "<arg_value>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154850": {
+      "content": "</arg_value>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154851": {
+      "content": "/nothink",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154852": {
+      "content": "<|begin_of_box|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154853": {
+      "content": "<|end_of_box|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154854": {
+      "content": "<|image|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    "154855": {
+      "content": "<|video|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "[MASK]",
+    "[gMASK]",
+    "[sMASK]",
+    "<sop>",
+    "<eop>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|observation|>",
+    "<|begin_of_image|>",
+    "<|end_of_image|>",
+    "<|begin_of_video|>",
+    "<|end_of_video|>",
+    "<|begin_of_audio|>",
+    "<|end_of_audio|>",
+    "<|begin_of_transcription|>",
+    "<|end_of_transcription|>"
+  ],
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 128000,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "remove_space": false,
+  "tokenizer_class": "PreTrainedTokenizer"
+}