Upload folder using huggingface_hub

chat_template.jinja

@@ -0,0 +1,87 @@
+{%- if messages[0]["role"] == "system" %}
+    {%- set system_message = messages[0]["content"] %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+{%- if not tools is defined %}
+    {%- set tools = none %}
+{%- endif %}
+{%- set user_messages = loop_messages | selectattr("role", "equalto", "user") | list %}
+
+{#- This block checks for alternating user/assistant messages, skipping tool calling messages #}
+{%- set ns = namespace() %}
+{%- set ns.index = 0 %}
+{%- for message in loop_messages %}
+    {%- if not (message.role == "tool" or message.role == "tool_results" or (message.tool_calls is defined and message.tool_calls is not none)) %}
+        {%- if (message["role"] == "user") != (ns.index % 2 == 0) %}
+            {{- raise_exception("After the optional system message, conversation roles must alternate user/assistant/user/assistant/...") }}
+        {%- endif %}
+        {%- set ns.index = ns.index + 1 %}
+    {%- endif %}
+{%- endfor %}
+
+{{- bos_token }}
+{%- for message in loop_messages %}
+    {%- if message["role"] == "user" %}
+        {%- if tools is not none and (message == user_messages[-1]) %}
+            {{- "[AVAILABLE_TOOLS] [" }}
+            {%- for tool in tools %}
+                {%- set tool = tool.function %}
+                {{- '{"type": "function", "function": {' }}
+                {%- for key, val in tool.items() if key != "return" %}
+                    {%- if val is string %}
+                        {{- '"' + key + '": "' + val + '"' }}
+                    {%- else %}
+                        {{- '"' + key + '": ' + val|tojson }}
+                    {%- endif %}
+                    {%- if not loop.last %}
+                        {{- ", " }}
+                    {%- endif %}
+                {%- endfor %}
+                {{- "}}" }}
+                {%- if not loop.last %}
+                    {{- ", " }}
+                {%- else %}
+                    {{- "]" }}
+                {%- endif %}
+            {%- endfor %}
+            {{- "[/AVAILABLE_TOOLS]" }}
+            {%- endif %}
+        {%- if loop.last and system_message is defined %}
+            {{- "[INST] " + system_message + "\n\n" + message["content"] + "[/INST]" }}
+        {%- else %}
+            {{- "[INST] " + message["content"] + "[/INST]" }}
+        {%- endif %}
+    {%- elif message.tool_calls is defined and message.tool_calls is not none %}
+        {{- "[TOOL_CALLS] [" }}
+        {%- for tool_call in message.tool_calls %}
+            {%- set out = tool_call.function|tojson %}
+            {{- out[:-1] }}
+            {%- if not tool_call.id is defined or tool_call.id|length != 9 %}
+                {{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }}
+            {%- endif %}
+            {{- ', "id": "' + tool_call.id + '"}' }}
+            {%- if not loop.last %}
+                {{- ", " }}
+            {%- else %}
+                {{- "]" + eos_token }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif message["role"] == "assistant" %}
+        {{- " " + message["content"]|trim + eos_token}}
+    {%- elif message["role"] == "tool_results" or message["role"] == "tool" %}
+        {%- if message.content is defined and message.content.content is defined %}
+            {%- set content = message.content.content %}
+        {%- else %}
+            {%- set content = message.content %}
+        {%- endif %}
+        {{- '[TOOL_RESULTS] {"content": ' + content|string + ", " }}
+        {%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %}
+            {{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }}
+        {%- endif %}
+        {{- '"call_id": "' + message.tool_call_id + '"}[/TOOL_RESULTS]' }}
+    {%- else %}
+        {{- raise_exception("Only user and assistant roles are supported, with the exception of an initial optional system message!") }}
+    {%- endif %}
+{%- endfor %}

config.json

@@ -0,0 +1,103 @@
+{
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "quantization_config": {
+    "bits": 8,
+    "checkpoint_format": "gptq",
+    "desc_act": false,
+    "dynamic": {
+      "+:.*\\.20\\..*": {
+        "bits": 4
+      },
+      "+:.*\\.23\\..*": {
+        "bits": 4
+      },
+      "+:.*\\.24\\..*": {
+        "bits": 4
+      },
+      "+:.*\\.25\\..*": {
+        "bits": 4
+      },
+      "+:.*\\.26\\..*": {
+        "bits": 4
+      },
+      "+:.*\\.27\\..*": {
+        "bits": 4
+      },
+      "+:.*\\.28\\..*": {
+        "bits": 4
+      },
+      "+:.*\\.29\\..*": {
+        "bits": 4
+      }
+    },
+    "format": "gptq",
+    "group_size": 128,
+    "lm_head": false,
+    "meta": {
+      "act_group_aware": true,
+      "auto_forward_data_parallel": true,
+      "damp_auto_increment": 0.01,
+      "damp_percent": 0.05,
+      "dense_vram_strategy": "exclusive",
+      "dense_vram_strategy_devices": null,
+      "fallback": {
+        "smooth": null,
+        "strategy": "rtn",
+        "threshold": "0.5%"
+      },
+      "foem": null,
+      "gc_mode": "interval",
+      "gptaq": null,
+      "hessian": {
+        "chunk_bytes": null,
+        "chunk_size": null,
+        "staging_dtype": "float32"
+      },
+      "mock_quantization": false,
+      "moe_vram_strategy": "exclusive",
+      "moe_vram_strategy_devices": null,
+      "mse": 0.0,
+      "offload_to_disk": true,
+      "offload_to_disk_path": "/tmp/gptqmodel_rlebkvo4",
+      "pack_impl": "cpu",
+      "quantizer": [
+        "gptqmodel:7.0.0"
+      ],
+      "static_groups": false,
+      "true_sequential": true,
+      "uri": "https://github.com/modelcloud/gptqmodel",
+      "wait_for_submodule_finalizers": false
+    },
+    "method": "gptq",
+    "pack_dtype": "int32",
+    "quant_method": "gptq",
+    "sym": true
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_parameters": {
+    "rope_theta": 1000000.0,
+    "rope_type": "default"
+  },
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "5.8.1",
+  "use_cache": true,
+  "vocab_size": 32768
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "transformers_version": "5.8.1"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31b3b9f6ca9f7af5f40a5e772993221c04d9e7018bdb402d93a1810d4caac9ec
+size 4275596832

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:071b832fd4d6db23c232d8b1525f278bc963130c61bc752029fb73ad213dff9f
+size 2530564488

model.safetensors.index.json

@@ -0,0 +1,970 @@
+{
+  "metadata": {
+    "total_size": 6806161320
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00001-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.g_idx": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.qweight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.qzeros": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.scales": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.scales": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.g_idx": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.qweight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.qzeros": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.scales": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00001-of-00002.safetensors"
+  }
+}

quant_log.csv

@@ -0,0 +1,225 @@
+layer,module,loss,samples,damp,time
+0,self_attn.k_proj,0.0000000045,0.05000,4.425
+0,self_attn.v_proj,0.0000000004,0.05000,4.681
+0,self_attn.q_proj,0.0000000150,0.05000,4.512
+0,self_attn.o_proj,0.0000000000,0.05000,1.517
+0,mlp.gate_proj,0.0000000164,0.05000,2.821
+0,mlp.up_proj,0.0000000142,0.05000,2.833
+0,mlp.down_proj,0.0000000000,0.05000,5.894
+1,self_attn.k_proj,0.0000000208,0.05000,4.715
+1,self_attn.v_proj,0.0000000034,0.05000,4.756
+1,self_attn.q_proj,0.0000000452,0.05000,4.779
+1,self_attn.o_proj,0.0000000000,0.05000,1.556
+1,mlp.up_proj,0.0000000480,0.05000,2.849
+1,mlp.gate_proj,0.0000000548,0.05000,2.858
+1,mlp.down_proj,0.0000000073,0.05000,5.973
+2,self_attn.v_proj,0.0000000225,0.05000,4.851
+2,self_attn.q_proj,0.0000003553,0.05000,4.918
+2,self_attn.k_proj,0.0000001725,0.05000,4.924
+2,self_attn.o_proj,0.0000000000,0.05000,1.544
+2,mlp.gate_proj,0.0000001153,0.05000,2.867
+2,mlp.up_proj,0.0000001006,0.05000,2.888
+2,mlp.down_proj,0.0000000000,0.05000,5.934
+3,self_attn.k_proj,0.0000000937,0.05000,4.973
+3,self_attn.v_proj,0.0000000155,0.05000,4.993
+3,self_attn.q_proj,0.0000001928,0.05000,5.032
+3,self_attn.o_proj,0.0000000000,0.05000,1.556
+3,mlp.up_proj,0.0000001660,0.05000,2.836
+3,mlp.gate_proj,0.0000001922,0.05000,2.846
+3,mlp.down_proj,0.0000000001,0.05000,5.986
+4,self_attn.k_proj,0.0000001324,0.05000,4.754
+4,self_attn.v_proj,0.0000000261,0.05000,4.777
+4,self_attn.q_proj,0.0000003002,0.05000,4.790
+4,self_attn.o_proj,0.0000000000,0.05000,1.549
+4,mlp.up_proj,0.0000002246,0.05000,2.851
+4,mlp.gate_proj,0.0000002774,0.05000,2.855
+4,mlp.down_proj,0.0000000001,0.05000,5.973
+5,self_attn.v_proj,0.0000000291,0.05000,4.803
+5,self_attn.k_proj,0.0000001730,0.05000,4.830
+5,self_attn.q_proj,0.0000003899,0.05000,4.837
+5,self_attn.o_proj,0.0000000001,0.05000,1.534
+5,mlp.up_proj,0.0000002912,0.05000,2.847
+5,mlp.gate_proj,0.0000003836,0.05000,2.857
+5,mlp.down_proj,0.0000000003,0.05000,5.888
+6,self_attn.v_proj,0.0000000294,0.05000,4.935
+6,self_attn.k_proj,0.0000001706,0.05000,4.948
+6,self_attn.q_proj,0.0000003636,0.05000,4.953
+6,self_attn.o_proj,0.0000000001,0.05000,1.521
+6,mlp.gate_proj,0.0000004451,0.05000,2.825
+6,mlp.up_proj,0.0000003460,0.05000,2.831
+6,mlp.down_proj,0.0000000004,0.05000,5.921
+7,self_attn.k_proj,0.0000002152,0.05000,4.900
+7,self_attn.q_proj,0.0000004503,0.05000,5.002
+7,self_attn.v_proj,0.0000000391,0.05000,5.020
+7,self_attn.o_proj,0.0000000001,0.05000,1.543
+7,mlp.up_proj,0.0000004005,0.05000,2.853
+7,mlp.gate_proj,0.0000005274,0.05000,2.873
+7,mlp.down_proj,0.0000000005,0.05000,5.870
+8,self_attn.k_proj,0.0000001697,0.05000,4.837
+8,self_attn.q_proj,0.0000003730,0.05000,4.889
+8,self_attn.v_proj,0.0000000364,0.05000,4.901
+8,self_attn.o_proj,0.0000000002,0.05000,1.528
+8,mlp.gate_proj,0.0000005689,0.05000,2.820
+8,mlp.up_proj,0.0000004404,0.05000,2.836
+8,mlp.down_proj,0.0000000005,0.05000,5.891
+9,self_attn.k_proj,0.0000002410,0.05000,4.779
+9,self_attn.v_proj,0.0000000435,0.05000,4.831
+9,self_attn.q_proj,0.0000005147,0.05000,4.858
+9,self_attn.o_proj,0.0000000002,0.05000,1.537
+9,mlp.up_proj,0.0000004793,0.05000,2.900
+9,mlp.gate_proj,0.0000005988,0.05000,2.916
+9,mlp.down_proj,0.0000000007,0.05000,5.878
+10,self_attn.v_proj,0.0000000379,0.05000,4.766
+10,self_attn.k_proj,0.0000002164,0.05000,4.832
+10,self_attn.q_proj,0.0000004548,0.05000,4.858
+10,self_attn.o_proj,0.0000000003,0.05000,1.525
+10,mlp.gate_proj,0.0000006227,0.05000,2.869
+10,mlp.up_proj,0.0000005106,0.05000,2.868
+10,mlp.down_proj,0.0000000008,0.05000,5.864
+11,self_attn.v_proj,0.0000000559,0.05000,4.834
+11,self_attn.k_proj,0.0000002533,0.05000,4.841
+11,self_attn.q_proj,0.0000005502,0.05000,4.852
+11,self_attn.o_proj,0.0000000004,0.05000,1.523
+11,mlp.up_proj,0.0000005648,0.05000,2.887
+11,mlp.gate_proj,0.0000006790,0.05000,2.893
+11,mlp.down_proj,0.0000000009,0.05000,5.876
+12,self_attn.v_proj,0.0000000635,0.05000,4.901
+12,self_attn.q_proj,0.0000007278,0.05000,4.935
+12,self_attn.k_proj,0.0000003287,0.05000,4.943
+12,self_attn.o_proj,0.0000000004,0.05000,1.547
+12,mlp.gate_proj,0.0000007479,0.05000,2.852
+12,mlp.up_proj,0.0000006364,0.05000,2.878
+12,mlp.down_proj,0.0000000011,0.05000,5.910
+13,self_attn.v_proj,0.0000000607,0.05000,4.963
+13,self_attn.k_proj,0.0000002807,0.05000,5.071
+13,self_attn.q_proj,0.0000005740,0.05000,5.114
+13,self_attn.o_proj,0.0000000006,0.05000,1.540
+13,mlp.up_proj,0.0000007320,0.05000,2.818
+13,mlp.gate_proj,0.0000008370,0.05000,2.828
+13,mlp.down_proj,0.0000000013,0.05000,5.928
+14,self_attn.v_proj,0.0000000978,0.05000,4.684
+14,self_attn.k_proj,0.0000002809,0.05000,4.698
+14,self_attn.q_proj,0.0000006548,0.05000,4.701
+14,self_attn.o_proj,0.0000000007,0.05000,1.589
+14,mlp.gate_proj,0.0000009377,0.05000,2.761
+14,mlp.up_proj,0.0000008157,0.05000,2.767
+14,mlp.down_proj,0.0000000017,0.05000,5.905
+15,self_attn.q_proj,0.0000008244,0.05000,4.845
+15,self_attn.k_proj,0.0000003675,0.05000,4.857
+15,self_attn.v_proj,0.0000001112,0.05000,4.876
+15,self_attn.o_proj,0.0000000008,0.05000,1.547
+15,mlp.up_proj,0.0000009010,0.05000,2.888
+15,mlp.gate_proj,0.0000010717,0.05000,2.894
+15,mlp.down_proj,0.0000000022,0.05000,5.838
+16,self_attn.q_proj,0.0000007387,0.05000,4.792
+16,self_attn.k_proj,0.0000003414,0.05000,4.793
+16,self_attn.v_proj,0.0000001060,0.05000,4.807
+16,self_attn.o_proj,0.0000000009,0.05000,1.543
+16,mlp.gate_proj,0.0000013285,0.05000,2.857
+16,mlp.up_proj,0.0000010571,0.05000,2.870
+16,mlp.down_proj,0.0000000031,0.05000,5.940
+17,self_attn.v_proj,0.0000001007,0.05000,4.856
+17,self_attn.k_proj,0.0000002850,0.05000,4.927
+17,self_attn.q_proj,0.0000006805,0.05000,4.947
+17,self_attn.o_proj,0.0000000011,0.05000,1.536
+17,mlp.up_proj,0.0000012283,0.05000,2.818
+17,mlp.gate_proj,0.0000015413,0.05000,2.826
+17,mlp.down_proj,0.0000000043,0.05000,5.896
+18,self_attn.v_proj,0.0000001238,0.05000,4.909
+18,self_attn.k_proj,0.0000003386,0.05000,4.934
+18,self_attn.q_proj,0.0000008637,0.05000,4.940
+18,self_attn.o_proj,0.0000000011,0.05000,1.552
+18,mlp.gate_proj,0.0000017367,0.05000,2.819
+18,mlp.up_proj,0.0000013842,0.05000,2.828
+18,mlp.down_proj,0.0000000061,0.05000,5.913
+19,self_attn.k_proj,0.0000003408,0.05000,4.792
+19,self_attn.v_proj,0.0000001450,0.05000,4.812
+19,self_attn.q_proj,0.0000007988,0.05000,4.818
+19,self_attn.o_proj,0.0000000013,0.05000,1.585
+19,mlp.up_proj,0.0000015377,0.05000,2.836
+19,mlp.gate_proj,0.0000019642,0.05000,2.841
+19,mlp.down_proj,0.0000000081,0.05000,5.875
+20,self_attn.k_proj,0.0001023333,0.05000,4.825
+20,self_attn.q_proj,0.0002501937,0.05000,4.835
+20,self_attn.v_proj,0.0000460283,0.05000,4.843
+20,self_attn.o_proj,0.0000003618,0.05000,1.527
+20,mlp.gate_proj,0.0006521194,0.05000,2.855
+20,mlp.up_proj,0.0004975359,0.05000,2.866
+20,mlp.down_proj,0.0000025065,0.05000,5.920
+21,self_attn.v_proj,0.0000001681,0.05000,4.679
+21,self_attn.k_proj,0.0000003462,0.05000,4.699
+21,self_attn.q_proj,0.0000008533,0.05000,4.729
+21,self_attn.o_proj,0.0000000012,0.05000,1.538
+21,mlp.gate_proj,0.0000026566,0.05000,2.859
+21,mlp.up_proj,0.0000018903,0.05000,2.859
+21,mlp.down_proj,0.0000000088,0.05000,5.901
+22,self_attn.v_proj,0.0000001750,0.05000,4.938
+22,self_attn.k_proj,0.0000003291,0.05000,5.021
+22,self_attn.q_proj,0.0000008261,0.05000,5.033
+22,self_attn.o_proj,0.0000000007,0.05000,1.538
+22,mlp.gate_proj,0.0000028447,0.05000,2.773
+22,mlp.up_proj,0.0000020450,0.05000,2.780
+22,mlp.down_proj,0.0000000093,0.05000,5.958
+23,self_attn.q_proj,0.0002417441,0.05000,4.984
+23,self_attn.k_proj,0.0000953232,0.05000,5.046
+23,self_attn.v_proj,0.0000531522,0.05000,5.053
+23,self_attn.o_proj,0.0000003526,0.05000,1.563
+23,mlp.up_proj,0.0006533562,0.05000,2.885
+23,mlp.gate_proj,0.0009118593,0.05000,2.905
+23,mlp.down_proj,0.0000030385,0.05000,6.019
+24,self_attn.k_proj,0.0001111512,0.05000,4.693
+24,self_attn.q_proj,0.0002774746,0.05000,4.699
+24,self_attn.v_proj,0.0000621167,0.05000,4.709
+24,self_attn.o_proj,0.0000003281,0.05000,1.552
+24,mlp.up_proj,0.0007241608,0.05000,2.855
+24,mlp.gate_proj,0.0010297146,0.05000,2.854
+24,mlp.down_proj,0.0000033343,0.05000,5.911
+25,self_attn.v_proj,0.0000711168,0.05000,4.587
+25,self_attn.k_proj,0.0001070380,0.05000,4.607
+25,self_attn.q_proj,0.0002784348,0.05000,4.614
+25,self_attn.o_proj,0.0000003775,0.05000,1.546
+25,mlp.gate_proj,0.0011190466,0.05000,2.868
+25,mlp.up_proj,0.0007940060,0.05000,2.875
+25,mlp.down_proj,0.0000038247,0.05000,5.850
+26,self_attn.v_proj,0.0000750327,0.05000,4.572
+26,self_attn.k_proj,0.0001003102,0.05000,4.604
+26,self_attn.q_proj,0.0002635715,0.05000,4.633
+26,self_attn.o_proj,0.0000004931,0.05000,1.531
+26,mlp.up_proj,0.0008625213,0.05000,2.902
+26,mlp.gate_proj,0.0011732855,0.05000,2.912
+26,mlp.down_proj,0.0000043874,0.05000,5.909
+27,self_attn.k_proj,0.0001011328,0.05000,4.689
+27,self_attn.q_proj,0.0002715807,0.05000,4.734
+27,self_attn.v_proj,0.0000679950,0.05000,4.742
+27,self_attn.o_proj,0.0000005592,0.05000,1.525
+27,mlp.up_proj,0.0009492542,0.05000,2.856
+27,mlp.gate_proj,0.0012830498,0.05000,2.880
+27,mlp.down_proj,0.0000051588,0.05000,5.885
+28,self_attn.v_proj,0.0000991559,0.05000,4.726
+28,self_attn.q_proj,0.0002645180,0.05000,4.752
+28,self_attn.k_proj,0.0001015245,0.05000,4.773
+28,self_attn.o_proj,0.0000008684,0.05000,1.541
+28,mlp.gate_proj,0.0013707465,0.05000,2.855
+28,mlp.up_proj,0.0010521358,0.05000,2.867
+28,mlp.down_proj,0.0000069794,0.05000,5.925
+29,self_attn.v_proj,0.0001437969,0.05000,4.577
+29,self_attn.k_proj,0.0001024157,0.05000,4.606
+29,self_attn.q_proj,0.0003022972,0.05000,4.619
+29,self_attn.o_proj,0.0000018183,0.05000,1.548
+29,mlp.gate_proj,0.0013315470,0.05000,2.857
+29,mlp.up_proj,0.0010652888,0.05000,2.869
+29,mlp.down_proj,0.0000088039,0.05000,5.885
+30,self_attn.v_proj,0.0000005459,0.05000,4.485
+30,self_attn.k_proj,0.0000003257,0.05000,4.520
+30,self_attn.q_proj,0.0000009694,0.05000,4.527
+30,self_attn.o_proj,0.0000000056,0.05000,1.534
+30,mlp.up_proj,0.0000037660,0.05000,2.806
+30,mlp.gate_proj,0.0000046258,0.05000,2.810
+30,mlp.down_proj,0.0000000387,0.05000,5.990
+31,self_attn.k_proj,0.0000003086,0.05000,4.720
+31,self_attn.v_proj,0.0000005535,0.05000,4.749
+31,self_attn.q_proj,0.0000008998,0.05000,4.763
+31,self_attn.o_proj,0.0000000074,0.05000,1.537
+31,mlp.gate_proj,0.0000037957,0.05000,2.888
+31,mlp.up_proj,0.0000029992,0.05000,2.888
+31,mlp.down_proj,0.0000000545,0.05000,5.847

quantize_config.json

@@ -0,0 +1,73 @@
+{
+  "bits": 8,
+  "dynamic": {
+    "+:.*\\.20\\..*": {
+      "bits": 4
+    },
+    "+:.*\\.23\\..*": {
+      "bits": 4
+    },
+    "+:.*\\.24\\..*": {
+      "bits": 4
+    },
+    "+:.*\\.25\\..*": {
+      "bits": 4
+    },
+    "+:.*\\.26\\..*": {
+      "bits": 4
+    },
+    "+:.*\\.27\\..*": {
+      "bits": 4
+    },
+    "+:.*\\.28\\..*": {
+      "bits": 4
+    },
+    "+:.*\\.29\\..*": {
+      "bits": 4
+    }
+  },
+  "group_size": 128,
+  "desc_act": false,
+  "lm_head": false,
+  "method": "gptq",
+  "quant_method": "gptq",
+  "format": "gptq",
+  "checkpoint_format": "gptq",
+  "pack_dtype": "int32",
+  "meta": {
+    "quantizer": [
+      "gptqmodel:7.0.0"
+    ],
+    "uri": "https://github.com/modelcloud/gptqmodel",
+    "damp_percent": 0.05,
+    "damp_auto_increment": 0.01,
+    "static_groups": false,
+    "true_sequential": true,
+    "mse": 0.0,
+    "gptaq": null,
+    "foem": null,
+    "act_group_aware": true,
+    "fallback": {
+      "strategy": "rtn",
+      "threshold": "0.5%",
+      "smooth": null
+    },
+    "offload_to_disk": true,
+    "offload_to_disk_path": "/tmp/gptqmodel_rlebkvo4",
+    "pack_impl": "cpu",
+    "gc_mode": "interval",
+    "wait_for_submodule_finalizers": false,
+    "auto_forward_data_parallel": true,
+    "dense_vram_strategy": "exclusive",
+    "dense_vram_strategy_devices": null,
+    "moe_vram_strategy": "exclusive",
+    "moe_vram_strategy_devices": null,
+    "mock_quantization": false,
+    "hessian": {
+      "chunk_size": null,
+      "chunk_bytes": null,
+      "staging_dtype": "float32"
+    }
+  },
+  "sym": true
+}

tokenizer_config.json

@@ -0,0 +1,17 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "is_local": false,
+  "legacy": false,
+  "local_files_only": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}