Upload folder using huggingface_hub

chat_template.jinja

@@ -1,87 +1,15 @@
-{%- 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 %}
+{% for message in messages %}
+{% if message['role'] == 'user' %}
+{{ '<|user|>
+' + message['content'] + eos_token }}
+{% elif message['role'] == 'system' %}
+{{ '<|system|>
+' + message['content'] + eos_token }}
+{% elif message['role'] == 'assistant' %}
+{{ '<|assistant|>
+'  + message['content'] + eos_token }}
+{% endif %}
+{% if loop.last and add_generation_prompt %}
+{{ '<|assistant|>' }}
+{% endif %}
+{% endfor %}

config.json

@@ -1,26 +1,29 @@
 {
   "architectures": [
-    "MistralForCausalLM"
+    "LlamaForCausalLM"
   ],
+  "attention_bias": false,
   "attention_dropout": 0.0,
   "bos_token_id": 1,
   "eos_token_id": 2,
-  "head_dim": null,
+  "head_dim": 64,
   "hidden_act": "silu",
-  "hidden_size": 4096,
+  "hidden_size": 2048,
   "initializer_range": 0.02,
-  "intermediate_size": 14336,
-  "max_position_embeddings": 32768,
-  "model_type": "mistral",
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "mlp_bias": false,
+  "model_type": "llama",
   "num_attention_heads": 32,
-  "num_hidden_layers": 32,
-  "num_key_value_heads": 8,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
   "rms_norm_eps": 1e-05,
-  "rope_theta": 1000000.0,
-  "sliding_window": null,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
   "tie_word_embeddings": false,
-  "torch_dtype": "bfloat16",
-  "transformers_version": "4.53.2",
+  "torch_dtype": "float32",
+  "transformers_version": "4.53.1",
   "use_cache": false,
-  "vocab_size": 32768
+  "vocab_size": 32000
 }

generation_config.json

@@ -1,6 +1,7 @@
 {
-  "_from_model_config": true,
   "bos_token_id": 1,
   "eos_token_id": 2,
-  "transformers_version": "4.53.2"
+  "max_length": 2048,
+  "pad_token_id": 0,
+  "transformers_version": "4.53.1"
 }

model-00001-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1e9422733d24ed227e344e1ce6581b7bd56d4353d269028acf70ee54ee6ea13
+oid sha256:b62321c11e762c8cc3bd96ddfb59b9191d63ff3a47ef0d9a5de912fd7749230c
 size 262144136

model-00002-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f2cb1590a68920a720bf8006b239e61cdc4ba3ae9d3f5a7eb346b611bf78807a
+oid sha256:bb91a5cb3ecc09d3f1d86e484910734f42e4b20fe0194d8fbcedda0c881bf62f
 size 197150040

model-00003-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a8bddd2212eca0450ee6b7b7c9628906bf902f1a5b30eae697771d090c2a4a4
+oid sha256:c511935eb1c916982d7c485d8e6c54ec68ea0cf7593cf0331931d3343c52f4e6
 size 192955512

model-00004-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7996f451ae0930a39099f22eb6842c94112dcf66195a114978b32516b2e4653
+oid sha256:b434a98e485c658c95dfddea4056392297df353c920562f8f60f29be01eb58da
 size 176178184

model-00005-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e4b81dba86b42837b4a465c5779a090ee634383c7c25fdf872fd7e4a8aeaaf9
+oid sha256:60c49ef47afba29036caf612d6bd245ab90ffaf144aec8aeaf22db3dad8ef400
 size 176178184

model-00006-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d486c9107072986c7d98be526c985e4e8bb5ecb3612e8e330545ff2f0d2ac6e
+oid sha256:846cd77c9e4b2577baa2e02caba2f46b8c5ada76d87bcc22de808170bb47a2e2
 size 176178184

model-00007-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fcaeb9d9df8fc2367aa1ca74f24e0d2245b18da9ebb612e3fc26dadbb0108f19
+oid sha256:649f13cf2154c46e790aaf46885eea73c99f2e5aa229cadc2343f2c43fc963d0
 size 176178184

model-00008-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:137def524e695a54962bb8bbece9c00ccb51007664a8afd659dfbb2eba7d5577
+oid sha256:98d7ee8ca2b430bd26e9f6cc8dd33fa2d74fc3c9c2e8a382efc4c3842ca5d799
 size 176178184

model-00009-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e8b0c8559ea6fe06e2d686d913d5cbb68b8f0271cecdd0fc5b96168c0281bd8
+oid sha256:d30b97b5ed8e9d72eab882161ba7697d01dd54dfad6370791d36d8823829ae5a
 size 176178184

model-00010-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8d11c76dd16bdc3f5caeddc4c6936e106e1ce90667d2f7b1736a1dfdd15e574
+oid sha256:13c28ecbb3987859704fe2d23e74252a85d081e0bc688b7df95ba21acd333ef9
 size 176178184

model-00011-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64d65fc7943294f731867f4b912bbd32f6d04108ae0d05d95a581576d4f58b3b
+oid sha256:eaece1e964711a0b0c375d70933fc4b7f6774f120c387a54d77bf989b444ad0e
 size 176178184

model-00012-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f94282f0716bb51899e52ac0c5ba49ef903d2912cc3c3bfe884016c71a67469a
+oid sha256:f7162fbad3ac944e66fb09d47b3acac44d10edac7c9876432b2a886d02dcf3e1
 size 176178192

model-00013-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9747ac9fe710a780e02a7639e9678f9306b4eacccb334a0f3221b033d34c4484
+oid sha256:077cbeabefbef764cba65c8082e92bfad8f90768fdd428572814e2d3a546644a
 size 176178192

model-00014-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd7b8d6c4026829d17300af5aabff1a6bb7a7d1affc08e67b8d4646df5a384f3
+oid sha256:379d2afa1ebbe12d8675b69d78b6b74c7163bcf7d4b3291b11bc4fbc1c445e89
 size 176178192

model-00015-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd8cc24fc431e66f6f3a5abedd7a197eeae153deb59ecb6299c598cc419598fa
+oid sha256:02f8ec42dec51741cf91d894a37bc3a63a8c593efe7d1f95e2151476891ec800
 size 176178192

model-00016-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3433cdf756d7cda471b89cb0d25610315cec619765659dce0a89b13fbeb16c02
+oid sha256:a732b9802e4e877d48b6a8211b451a5ad6eff43df973cb532cc3deb460311639
 size 176178192

model-00017-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edccb601ccf002e0ee7feb9b6796e97e3ee8e1bd8b962ae4172e6f03cba8ad6b
+oid sha256:9b0849044681ab876e593a584c557b1236b1262171ff3f261c2b032da9efb6f9
 size 176178192

model-00018-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97def02f851ef50d1861a5da1732f38bfafc01de05c8292e95d0fe160889959c
+oid sha256:a638437dbb7bca2007fa01cf3d87dd702cd8a4b703d1299fb3a0b0a35622508b
 size 176178192

model-00019-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b453cd4212aa0110b2ad02b8c7f1ba545f5ac545279d9d0fe999bd5353d9d42f
+oid sha256:0955b13347b65cf7c4377ee5f90366aaf8c58f83e6912b6ceca141fd821c69ef
 size 176178192

model-00020-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d80eb16e86f3c88046fa078262398d78787727e0239de0ab667e4ccf60f8c62e
+oid sha256:0d103f5801230cbf4fc11ddc3253b8c8227a207bcaf9b4d0278eb35d39b5f619
 size 176178192

model-00021-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:494b9d357e086a561420beed3312b105cdf6356f8454a2d9e758d0ec9049ca87
+oid sha256:9fb36a8eb1f94862d8d111f56a7c8c54355da621085dbf3dff03c7e6c32b4a3f
 size 176178192

model-00022-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60e44d63be4725e9c302d9e26709f2a46e270cf3efff2842d1818f62330b250c
+oid sha256:c371dfeb1a1b75b22379900d8e8a6098e5663e099aa322461789c8da244b2cc3
 size 176178192

model-00023-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41f807c6840f3e671c9f51f8137dc53041dae2a299862684f333185c2a2c7e17
+oid sha256:f2eaf564ac8e2e04e03222af0ac34d86e15dd0b7a24ef7e5739cc7165f1a9fdf
 size 262144128

model-00024-of-00024.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c4f8808fc7e793e47d4eca20a34d2a8dc5b2731a7a9a5b960f23def126cc11ad
+oid sha256:c6085625471dfc8fb2ef7f54c1cf2606de39894fca11e091a1dedc76f904e5ac
 size 138437280

model.safetensors.index.json

@@ -1,299 +1,209 @@
 {
   "metadata": {
-    "total_parameters": 7248023552,
-    "total_size": 14496047104
+    "total_parameters": 1100048384,
+    "total_size": 4400193536
   },
   "weight_map": {
-    "lm_head.weight": "model-00098-of-00099.safetensors",
-    "model.embed_tokens.weight": "model-00001-of-00099.safetensors",
-    "model.layers.0.input_layernorm.weight": "model-00005-of-00099.safetensors",
-    "model.layers.0.mlp.down_proj.weight": "model-00005-of-00099.safetensors",
-    "model.layers.0.mlp.gate_proj.weight": "model-00003-of-00099.safetensors",
-    "model.layers.0.mlp.up_proj.weight": "model-00004-of-00099.safetensors",
-    "model.layers.0.post_attention_layernorm.weight": "model-00005-of-00099.safetensors",
-    "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00099.safetensors",
-    "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00099.safetensors",
-    "model.layers.0.self_attn.q_proj.weight": "model-00002-of-00099.safetensors",
-    "model.layers.0.self_attn.v_proj.weight": "model-00002-of-00099.safetensors",
-    "model.layers.1.input_layernorm.weight": "model-00008-of-00099.safetensors",
-    "model.layers.1.mlp.down_proj.weight": "model-00008-of-00099.safetensors",
-    "model.layers.1.mlp.gate_proj.weight": "model-00006-of-00099.safetensors",
-    "model.layers.1.mlp.up_proj.weight": "model-00007-of-00099.safetensors",
-    "model.layers.1.post_attention_layernorm.weight": "model-00008-of-00099.safetensors",
-    "model.layers.1.self_attn.k_proj.weight": "model-00005-of-00099.safetensors",
-    "model.layers.1.self_attn.o_proj.weight": "model-00006-of-00099.safetensors",
-    "model.layers.1.self_attn.q_proj.weight": "model-00005-of-00099.safetensors",
-    "model.layers.1.self_attn.v_proj.weight": "model-00005-of-00099.safetensors",
-    "model.layers.10.input_layernorm.weight": "model-00035-of-00099.safetensors",
-    "model.layers.10.mlp.down_proj.weight": "model-00035-of-00099.safetensors",
-    "model.layers.10.mlp.gate_proj.weight": "model-00033-of-00099.safetensors",
-    "model.layers.10.mlp.up_proj.weight": "model-00034-of-00099.safetensors",
-    "model.layers.10.post_attention_layernorm.weight": "model-00035-of-00099.safetensors",
-    "model.layers.10.self_attn.k_proj.weight": "model-00032-of-00099.safetensors",
-    "model.layers.10.self_attn.o_proj.weight": "model-00033-of-00099.safetensors",
-    "model.layers.10.self_attn.q_proj.weight": "model-00032-of-00099.safetensors",
-    "model.layers.10.self_attn.v_proj.weight": "model-00032-of-00099.safetensors",
-    "model.layers.11.input_layernorm.weight": "model-00038-of-00099.safetensors",
-    "model.layers.11.mlp.down_proj.weight": "model-00038-of-00099.safetensors",
-    "model.layers.11.mlp.gate_proj.weight": "model-00036-of-00099.safetensors",
-    "model.layers.11.mlp.up_proj.weight": "model-00037-of-00099.safetensors",
-    "model.layers.11.post_attention_layernorm.weight": "model-00038-of-00099.safetensors",
-    "model.layers.11.self_attn.k_proj.weight": "model-00035-of-00099.safetensors",
-    "model.layers.11.self_attn.o_proj.weight": "model-00036-of-00099.safetensors",
-    "model.layers.11.self_attn.q_proj.weight": "model-00035-of-00099.safetensors",
-    "model.layers.11.self_attn.v_proj.weight": "model-00035-of-00099.safetensors",
-    "model.layers.12.input_layernorm.weight": "model-00041-of-00099.safetensors",
-    "model.layers.12.mlp.down_proj.weight": "model-00041-of-00099.safetensors",
-    "model.layers.12.mlp.gate_proj.weight": "model-00039-of-00099.safetensors",
-    "model.layers.12.mlp.up_proj.weight": "model-00040-of-00099.safetensors",
-    "model.layers.12.post_attention_layernorm.weight": "model-00041-of-00099.safetensors",
-    "model.layers.12.self_attn.k_proj.weight": "model-00038-of-00099.safetensors",
-    "model.layers.12.self_attn.o_proj.weight": "model-00039-of-00099.safetensors",
-    "model.layers.12.self_attn.q_proj.weight": "model-00038-of-00099.safetensors",
-    "model.layers.12.self_attn.v_proj.weight": "model-00038-of-00099.safetensors",
-    "model.layers.13.input_layernorm.weight": "model-00044-of-00099.safetensors",
-    "model.layers.13.mlp.down_proj.weight": "model-00044-of-00099.safetensors",
-    "model.layers.13.mlp.gate_proj.weight": "model-00042-of-00099.safetensors",
-    "model.layers.13.mlp.up_proj.weight": "model-00043-of-00099.safetensors",
-    "model.layers.13.post_attention_layernorm.weight": "model-00044-of-00099.safetensors",
-    "model.layers.13.self_attn.k_proj.weight": "model-00041-of-00099.safetensors",
-    "model.layers.13.self_attn.o_proj.weight": "model-00042-of-00099.safetensors",
-    "model.layers.13.self_attn.q_proj.weight": "model-00041-of-00099.safetensors",
-    "model.layers.13.self_attn.v_proj.weight": "model-00041-of-00099.safetensors",
-    "model.layers.14.input_layernorm.weight": "model-00047-of-00099.safetensors",
-    "model.layers.14.mlp.down_proj.weight": "model-00047-of-00099.safetensors",
-    "model.layers.14.mlp.gate_proj.weight": "model-00045-of-00099.safetensors",
-    "model.layers.14.mlp.up_proj.weight": "model-00046-of-00099.safetensors",
-    "model.layers.14.post_attention_layernorm.weight": "model-00047-of-00099.safetensors",
-    "model.layers.14.self_attn.k_proj.weight": "model-00044-of-00099.safetensors",
-    "model.layers.14.self_attn.o_proj.weight": "model-00045-of-00099.safetensors",
-    "model.layers.14.self_attn.q_proj.weight": "model-00044-of-00099.safetensors",
-    "model.layers.14.self_attn.v_proj.weight": "model-00044-of-00099.safetensors",
-    "model.layers.15.input_layernorm.weight": "model-00050-of-00099.safetensors",
-    "model.layers.15.mlp.down_proj.weight": "model-00050-of-00099.safetensors",
-    "model.layers.15.mlp.gate_proj.weight": "model-00048-of-00099.safetensors",
-    "model.layers.15.mlp.up_proj.weight": "model-00049-of-00099.safetensors",
-    "model.layers.15.post_attention_layernorm.weight": "model-00050-of-00099.safetensors",
-    "model.layers.15.self_attn.k_proj.weight": "model-00047-of-00099.safetensors",
-    "model.layers.15.self_attn.o_proj.weight": "model-00048-of-00099.safetensors",
-    "model.layers.15.self_attn.q_proj.weight": "model-00047-of-00099.safetensors",
-    "model.layers.15.self_attn.v_proj.weight": "model-00047-of-00099.safetensors",
-    "model.layers.16.input_layernorm.weight": "model-00053-of-00099.safetensors",
-    "model.layers.16.mlp.down_proj.weight": "model-00053-of-00099.safetensors",
-    "model.layers.16.mlp.gate_proj.weight": "model-00051-of-00099.safetensors",
-    "model.layers.16.mlp.up_proj.weight": "model-00052-of-00099.safetensors",
-    "model.layers.16.post_attention_layernorm.weight": "model-00053-of-00099.safetensors",
-    "model.layers.16.self_attn.k_proj.weight": "model-00050-of-00099.safetensors",
-    "model.layers.16.self_attn.o_proj.weight": "model-00051-of-00099.safetensors",
-    "model.layers.16.self_attn.q_proj.weight": "model-00050-of-00099.safetensors",
-    "model.layers.16.self_attn.v_proj.weight": "model-00050-of-00099.safetensors",
-    "model.layers.17.input_layernorm.weight": "model-00056-of-00099.safetensors",
-    "model.layers.17.mlp.down_proj.weight": "model-00056-of-00099.safetensors",
-    "model.layers.17.mlp.gate_proj.weight": "model-00054-of-00099.safetensors",
-    "model.layers.17.mlp.up_proj.weight": "model-00055-of-00099.safetensors",
-    "model.layers.17.post_attention_layernorm.weight": "model-00056-of-00099.safetensors",
-    "model.layers.17.self_attn.k_proj.weight": "model-00053-of-00099.safetensors",
-    "model.layers.17.self_attn.o_proj.weight": "model-00054-of-00099.safetensors",
-    "model.layers.17.self_attn.q_proj.weight": "model-00053-of-00099.safetensors",
-    "model.layers.17.self_attn.v_proj.weight": "model-00053-of-00099.safetensors",
-    "model.layers.18.input_layernorm.weight": "model-00059-of-00099.safetensors",
-    "model.layers.18.mlp.down_proj.weight": "model-00059-of-00099.safetensors",
-    "model.layers.18.mlp.gate_proj.weight": "model-00057-of-00099.safetensors",
-    "model.layers.18.mlp.up_proj.weight": "model-00058-of-00099.safetensors",
-    "model.layers.18.post_attention_layernorm.weight": "model-00059-of-00099.safetensors",
-    "model.layers.18.self_attn.k_proj.weight": "model-00056-of-00099.safetensors",
-    "model.layers.18.self_attn.o_proj.weight": "model-00057-of-00099.safetensors",
-    "model.layers.18.self_attn.q_proj.weight": "model-00056-of-00099.safetensors",
-    "model.layers.18.self_attn.v_proj.weight": "model-00056-of-00099.safetensors",
-    "model.layers.19.input_layernorm.weight": "model-00062-of-00099.safetensors",
-    "model.layers.19.mlp.down_proj.weight": "model-00062-of-00099.safetensors",
-    "model.layers.19.mlp.gate_proj.weight": "model-00060-of-00099.safetensors",
-    "model.layers.19.mlp.up_proj.weight": "model-00061-of-00099.safetensors",
-    "model.layers.19.post_attention_layernorm.weight": "model-00062-of-00099.safetensors",
-    "model.layers.19.self_attn.k_proj.weight": "model-00059-of-00099.safetensors",
-    "model.layers.19.self_attn.o_proj.weight": "model-00060-of-00099.safetensors",
-    "model.layers.19.self_attn.q_proj.weight": "model-00059-of-00099.safetensors",
-    "model.layers.19.self_attn.v_proj.weight": "model-00059-of-00099.safetensors",
-    "model.layers.2.input_layernorm.weight": "model-00011-of-00099.safetensors",
-    "model.layers.2.mlp.down_proj.weight": "model-00011-of-00099.safetensors",
-    "model.layers.2.mlp.gate_proj.weight": "model-00009-of-00099.safetensors",
-    "model.layers.2.mlp.up_proj.weight": "model-00010-of-00099.safetensors",
-    "model.layers.2.post_attention_layernorm.weight": "model-00011-of-00099.safetensors",
-    "model.layers.2.self_attn.k_proj.weight": "model-00008-of-00099.safetensors",
-    "model.layers.2.self_attn.o_proj.weight": "model-00009-of-00099.safetensors",
-    "model.layers.2.self_attn.q_proj.weight": "model-00008-of-00099.safetensors",
-    "model.layers.2.self_attn.v_proj.weight": "model-00008-of-00099.safetensors",
-    "model.layers.20.input_layernorm.weight": "model-00065-of-00099.safetensors",
-    "model.layers.20.mlp.down_proj.weight": "model-00065-of-00099.safetensors",
-    "model.layers.20.mlp.gate_proj.weight": "model-00063-of-00099.safetensors",
-    "model.layers.20.mlp.up_proj.weight": "model-00064-of-00099.safetensors",
-    "model.layers.20.post_attention_layernorm.weight": "model-00065-of-00099.safetensors",
-    "model.layers.20.self_attn.k_proj.weight": "model-00062-of-00099.safetensors",
-    "model.layers.20.self_attn.o_proj.weight": "model-00063-of-00099.safetensors",
-    "model.layers.20.self_attn.q_proj.weight": "model-00062-of-00099.safetensors",
-    "model.layers.20.self_attn.v_proj.weight": "model-00062-of-00099.safetensors",
-    "model.layers.21.input_layernorm.weight": "model-00068-of-00099.safetensors",
-    "model.layers.21.mlp.down_proj.weight": "model-00068-of-00099.safetensors",
-    "model.layers.21.mlp.gate_proj.weight": "model-00066-of-00099.safetensors",
-    "model.layers.21.mlp.up_proj.weight": "model-00067-of-00099.safetensors",
-    "model.layers.21.post_attention_layernorm.weight": "model-00068-of-00099.safetensors",
-    "model.layers.21.self_attn.k_proj.weight": "model-00065-of-00099.safetensors",
-    "model.layers.21.self_attn.o_proj.weight": "model-00066-of-00099.safetensors",
-    "model.layers.21.self_attn.q_proj.weight": "model-00065-of-00099.safetensors",
-    "model.layers.21.self_attn.v_proj.weight": "model-00065-of-00099.safetensors",
-    "model.layers.22.input_layernorm.weight": "model-00071-of-00099.safetensors",
-    "model.layers.22.mlp.down_proj.weight": "model-00071-of-00099.safetensors",
-    "model.layers.22.mlp.gate_proj.weight": "model-00069-of-00099.safetensors",
-    "model.layers.22.mlp.up_proj.weight": "model-00070-of-00099.safetensors",
-    "model.layers.22.post_attention_layernorm.weight": "model-00071-of-00099.safetensors",
-    "model.layers.22.self_attn.k_proj.weight": "model-00068-of-00099.safetensors",
-    "model.layers.22.self_attn.o_proj.weight": "model-00069-of-00099.safetensors",
-    "model.layers.22.self_attn.q_proj.weight": "model-00068-of-00099.safetensors",
-    "model.layers.22.self_attn.v_proj.weight": "model-00068-of-00099.safetensors",
-    "model.layers.23.input_layernorm.weight": "model-00074-of-00099.safetensors",
-    "model.layers.23.mlp.down_proj.weight": "model-00074-of-00099.safetensors",
-    "model.layers.23.mlp.gate_proj.weight": "model-00072-of-00099.safetensors",
-    "model.layers.23.mlp.up_proj.weight": "model-00073-of-00099.safetensors",
-    "model.layers.23.post_attention_layernorm.weight": "model-00074-of-00099.safetensors",
-    "model.layers.23.self_attn.k_proj.weight": "model-00071-of-00099.safetensors",
-    "model.layers.23.self_attn.o_proj.weight": "model-00072-of-00099.safetensors",
-    "model.layers.23.self_attn.q_proj.weight": "model-00071-of-00099.safetensors",
-    "model.layers.23.self_attn.v_proj.weight": "model-00071-of-00099.safetensors",
-    "model.layers.24.input_layernorm.weight": "model-00077-of-00099.safetensors",
-    "model.layers.24.mlp.down_proj.weight": "model-00077-of-00099.safetensors",
-    "model.layers.24.mlp.gate_proj.weight": "model-00075-of-00099.safetensors",
-    "model.layers.24.mlp.up_proj.weight": "model-00076-of-00099.safetensors",
-    "model.layers.24.post_attention_layernorm.weight": "model-00077-of-00099.safetensors",
-    "model.layers.24.self_attn.k_proj.weight": "model-00074-of-00099.safetensors",
-    "model.layers.24.self_attn.o_proj.weight": "model-00075-of-00099.safetensors",
-    "model.layers.24.self_attn.q_proj.weight": "model-00074-of-00099.safetensors",
-    "model.layers.24.self_attn.v_proj.weight": "model-00074-of-00099.safetensors",
-    "model.layers.25.input_layernorm.weight": "model-00080-of-00099.safetensors",
-    "model.layers.25.mlp.down_proj.weight": "model-00080-of-00099.safetensors",
-    "model.layers.25.mlp.gate_proj.weight": "model-00078-of-00099.safetensors",
-    "model.layers.25.mlp.up_proj.weight": "model-00079-of-00099.safetensors",
-    "model.layers.25.post_attention_layernorm.weight": "model-00080-of-00099.safetensors",
-    "model.layers.25.self_attn.k_proj.weight": "model-00077-of-00099.safetensors",
-    "model.layers.25.self_attn.o_proj.weight": "model-00078-of-00099.safetensors",
-    "model.layers.25.self_attn.q_proj.weight": "model-00077-of-00099.safetensors",
-    "model.layers.25.self_attn.v_proj.weight": "model-00077-of-00099.safetensors",
-    "model.layers.26.input_layernorm.weight": "model-00083-of-00099.safetensors",
-    "model.layers.26.mlp.down_proj.weight": "model-00083-of-00099.safetensors",
-    "model.layers.26.mlp.gate_proj.weight": "model-00081-of-00099.safetensors",
-    "model.layers.26.mlp.up_proj.weight": "model-00082-of-00099.safetensors",
-    "model.layers.26.post_attention_layernorm.weight": "model-00083-of-00099.safetensors",
-    "model.layers.26.self_attn.k_proj.weight": "model-00080-of-00099.safetensors",
-    "model.layers.26.self_attn.o_proj.weight": "model-00081-of-00099.safetensors",
-    "model.layers.26.self_attn.q_proj.weight": "model-00080-of-00099.safetensors",
-    "model.layers.26.self_attn.v_proj.weight": "model-00080-of-00099.safetensors",
-    "model.layers.27.input_layernorm.weight": "model-00086-of-00099.safetensors",
-    "model.layers.27.mlp.down_proj.weight": "model-00086-of-00099.safetensors",
-    "model.layers.27.mlp.gate_proj.weight": "model-00084-of-00099.safetensors",
-    "model.layers.27.mlp.up_proj.weight": "model-00085-of-00099.safetensors",
-    "model.layers.27.post_attention_layernorm.weight": "model-00086-of-00099.safetensors",
-    "model.layers.27.self_attn.k_proj.weight": "model-00083-of-00099.safetensors",
-    "model.layers.27.self_attn.o_proj.weight": "model-00084-of-00099.safetensors",
-    "model.layers.27.self_attn.q_proj.weight": "model-00083-of-00099.safetensors",
-    "model.layers.27.self_attn.v_proj.weight": "model-00083-of-00099.safetensors",
-    "model.layers.28.input_layernorm.weight": "model-00089-of-00099.safetensors",
-    "model.layers.28.mlp.down_proj.weight": "model-00089-of-00099.safetensors",
-    "model.layers.28.mlp.gate_proj.weight": "model-00087-of-00099.safetensors",
-    "model.layers.28.mlp.up_proj.weight": "model-00088-of-00099.safetensors",
-    "model.layers.28.post_attention_layernorm.weight": "model-00089-of-00099.safetensors",
-    "model.layers.28.self_attn.k_proj.weight": "model-00086-of-00099.safetensors",
-    "model.layers.28.self_attn.o_proj.weight": "model-00087-of-00099.safetensors",
-    "model.layers.28.self_attn.q_proj.weight": "model-00086-of-00099.safetensors",
-    "model.layers.28.self_attn.v_proj.weight": "model-00086-of-00099.safetensors",
-    "model.layers.29.input_layernorm.weight": "model-00092-of-00099.safetensors",
-    "model.layers.29.mlp.down_proj.weight": "model-00092-of-00099.safetensors",
-    "model.layers.29.mlp.gate_proj.weight": "model-00090-of-00099.safetensors",
-    "model.layers.29.mlp.up_proj.weight": "model-00091-of-00099.safetensors",
-    "model.layers.29.post_attention_layernorm.weight": "model-00092-of-00099.safetensors",
-    "model.layers.29.self_attn.k_proj.weight": "model-00089-of-00099.safetensors",
-    "model.layers.29.self_attn.o_proj.weight": "model-00090-of-00099.safetensors",
-    "model.layers.29.self_attn.q_proj.weight": "model-00089-of-00099.safetensors",
-    "model.layers.29.self_attn.v_proj.weight": "model-00089-of-00099.safetensors",
-    "model.layers.3.input_layernorm.weight": "model-00014-of-00099.safetensors",
-    "model.layers.3.mlp.down_proj.weight": "model-00014-of-00099.safetensors",
-    "model.layers.3.mlp.gate_proj.weight": "model-00012-of-00099.safetensors",
-    "model.layers.3.mlp.up_proj.weight": "model-00013-of-00099.safetensors",
-    "model.layers.3.post_attention_layernorm.weight": "model-00014-of-00099.safetensors",
-    "model.layers.3.self_attn.k_proj.weight": "model-00011-of-00099.safetensors",
-    "model.layers.3.self_attn.o_proj.weight": "model-00012-of-00099.safetensors",
-    "model.layers.3.self_attn.q_proj.weight": "model-00011-of-00099.safetensors",
-    "model.layers.3.self_attn.v_proj.weight": "model-00011-of-00099.safetensors",
-    "model.layers.30.input_layernorm.weight": "model-00095-of-00099.safetensors",
-    "model.layers.30.mlp.down_proj.weight": "model-00095-of-00099.safetensors",
-    "model.layers.30.mlp.gate_proj.weight": "model-00093-of-00099.safetensors",
-    "model.layers.30.mlp.up_proj.weight": "model-00094-of-00099.safetensors",
-    "model.layers.30.post_attention_layernorm.weight": "model-00095-of-00099.safetensors",
-    "model.layers.30.self_attn.k_proj.weight": "model-00092-of-00099.safetensors",
-    "model.layers.30.self_attn.o_proj.weight": "model-00093-of-00099.safetensors",
-    "model.layers.30.self_attn.q_proj.weight": "model-00092-of-00099.safetensors",
-    "model.layers.30.self_attn.v_proj.weight": "model-00092-of-00099.safetensors",
-    "model.layers.31.input_layernorm.weight": "model-00099-of-00099.safetensors",
-    "model.layers.31.mlp.down_proj.weight": "model-00099-of-00099.safetensors",
-    "model.layers.31.mlp.gate_proj.weight": "model-00096-of-00099.safetensors",
-    "model.layers.31.mlp.up_proj.weight": "model-00097-of-00099.safetensors",
-    "model.layers.31.post_attention_layernorm.weight": "model-00099-of-00099.safetensors",
-    "model.layers.31.self_attn.k_proj.weight": "model-00095-of-00099.safetensors",
-    "model.layers.31.self_attn.o_proj.weight": "model-00096-of-00099.safetensors",
-    "model.layers.31.self_attn.q_proj.weight": "model-00095-of-00099.safetensors",
-    "model.layers.31.self_attn.v_proj.weight": "model-00095-of-00099.safetensors",
-    "model.layers.4.input_layernorm.weight": "model-00017-of-00099.safetensors",
-    "model.layers.4.mlp.down_proj.weight": "model-00017-of-00099.safetensors",
-    "model.layers.4.mlp.gate_proj.weight": "model-00015-of-00099.safetensors",
-    "model.layers.4.mlp.up_proj.weight": "model-00016-of-00099.safetensors",
-    "model.layers.4.post_attention_layernorm.weight": "model-00017-of-00099.safetensors",
-    "model.layers.4.self_attn.k_proj.weight": "model-00014-of-00099.safetensors",
-    "model.layers.4.self_attn.o_proj.weight": "model-00015-of-00099.safetensors",
-    "model.layers.4.self_attn.q_proj.weight": "model-00014-of-00099.safetensors",
-    "model.layers.4.self_attn.v_proj.weight": "model-00014-of-00099.safetensors",
-    "model.layers.5.input_layernorm.weight": "model-00020-of-00099.safetensors",
-    "model.layers.5.mlp.down_proj.weight": "model-00020-of-00099.safetensors",
-    "model.layers.5.mlp.gate_proj.weight": "model-00018-of-00099.safetensors",
-    "model.layers.5.mlp.up_proj.weight": "model-00019-of-00099.safetensors",
-    "model.layers.5.post_attention_layernorm.weight": "model-00020-of-00099.safetensors",
-    "model.layers.5.self_attn.k_proj.weight": "model-00017-of-00099.safetensors",
-    "model.layers.5.self_attn.o_proj.weight": "model-00018-of-00099.safetensors",
-    "model.layers.5.self_attn.q_proj.weight": "model-00017-of-00099.safetensors",
-    "model.layers.5.self_attn.v_proj.weight": "model-00017-of-00099.safetensors",
-    "model.layers.6.input_layernorm.weight": "model-00023-of-00099.safetensors",
-    "model.layers.6.mlp.down_proj.weight": "model-00023-of-00099.safetensors",
-    "model.layers.6.mlp.gate_proj.weight": "model-00021-of-00099.safetensors",
-    "model.layers.6.mlp.up_proj.weight": "model-00022-of-00099.safetensors",
-    "model.layers.6.post_attention_layernorm.weight": "model-00023-of-00099.safetensors",
-    "model.layers.6.self_attn.k_proj.weight": "model-00020-of-00099.safetensors",
-    "model.layers.6.self_attn.o_proj.weight": "model-00021-of-00099.safetensors",
-    "model.layers.6.self_attn.q_proj.weight": "model-00020-of-00099.safetensors",
-    "model.layers.6.self_attn.v_proj.weight": "model-00020-of-00099.safetensors",
-    "model.layers.7.input_layernorm.weight": "model-00026-of-00099.safetensors",
-    "model.layers.7.mlp.down_proj.weight": "model-00026-of-00099.safetensors",
-    "model.layers.7.mlp.gate_proj.weight": "model-00024-of-00099.safetensors",
-    "model.layers.7.mlp.up_proj.weight": "model-00025-of-00099.safetensors",
-    "model.layers.7.post_attention_layernorm.weight": "model-00026-of-00099.safetensors",
-    "model.layers.7.self_attn.k_proj.weight": "model-00023-of-00099.safetensors",
-    "model.layers.7.self_attn.o_proj.weight": "model-00024-of-00099.safetensors",
-    "model.layers.7.self_attn.q_proj.weight": "model-00023-of-00099.safetensors",
-    "model.layers.7.self_attn.v_proj.weight": "model-00023-of-00099.safetensors",
-    "model.layers.8.input_layernorm.weight": "model-00029-of-00099.safetensors",
-    "model.layers.8.mlp.down_proj.weight": "model-00029-of-00099.safetensors",
-    "model.layers.8.mlp.gate_proj.weight": "model-00027-of-00099.safetensors",
-    "model.layers.8.mlp.up_proj.weight": "model-00028-of-00099.safetensors",
-    "model.layers.8.post_attention_layernorm.weight": "model-00029-of-00099.safetensors",
-    "model.layers.8.self_attn.k_proj.weight": "model-00026-of-00099.safetensors",
-    "model.layers.8.self_attn.o_proj.weight": "model-00027-of-00099.safetensors",
-    "model.layers.8.self_attn.q_proj.weight": "model-00026-of-00099.safetensors",
-    "model.layers.8.self_attn.v_proj.weight": "model-00026-of-00099.safetensors",
-    "model.layers.9.input_layernorm.weight": "model-00032-of-00099.safetensors",
-    "model.layers.9.mlp.down_proj.weight": "model-00032-of-00099.safetensors",
-    "model.layers.9.mlp.gate_proj.weight": "model-00030-of-00099.safetensors",
-    "model.layers.9.mlp.up_proj.weight": "model-00031-of-00099.safetensors",
-    "model.layers.9.post_attention_layernorm.weight": "model-00032-of-00099.safetensors",
-    "model.layers.9.self_attn.k_proj.weight": "model-00029-of-00099.safetensors",
-    "model.layers.9.self_attn.o_proj.weight": "model-00030-of-00099.safetensors",
-    "model.layers.9.self_attn.q_proj.weight": "model-00029-of-00099.safetensors",
-    "model.layers.9.self_attn.v_proj.weight": "model-00029-of-00099.safetensors",
-    "model.norm.weight": "model-00099-of-00099.safetensors"
+    "lm_head.weight": "model-00023-of-00024.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00024.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00002-of-00024.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00024.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00003-of-00024.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00003-of-00024.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00003-of-00024.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00003-of-00024.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00003-of-00024.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00003-of-00024.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00024.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00012-of-00024.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00012-of-00024.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00012-of-00024.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00012-of-00024.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00012-of-00024.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00011-of-00024.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00011-of-00024.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00011-of-00024.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00011-of-00024.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00013-of-00024.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00013-of-00024.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00013-of-00024.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00013-of-00024.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00013-of-00024.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00012-of-00024.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00012-of-00024.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00012-of-00024.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00012-of-00024.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00014-of-00024.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00014-of-00024.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00014-of-00024.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00014-of-00024.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00014-of-00024.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00013-of-00024.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00013-of-00024.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00013-of-00024.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00013-of-00024.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00015-of-00024.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00015-of-00024.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00015-of-00024.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00015-of-00024.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00015-of-00024.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00014-of-00024.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00014-of-00024.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00014-of-00024.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00014-of-00024.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00016-of-00024.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00016-of-00024.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00016-of-00024.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00016-of-00024.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00016-of-00024.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00015-of-00024.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00015-of-00024.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00015-of-00024.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00015-of-00024.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00017-of-00024.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00017-of-00024.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00017-of-00024.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00017-of-00024.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00017-of-00024.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00016-of-00024.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00016-of-00024.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00016-of-00024.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00016-of-00024.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00018-of-00024.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00018-of-00024.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00018-of-00024.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00018-of-00024.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00018-of-00024.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00017-of-00024.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00017-of-00024.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00017-of-00024.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00017-of-00024.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00019-of-00024.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00019-of-00024.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00019-of-00024.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00019-of-00024.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00019-of-00024.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00018-of-00024.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00018-of-00024.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00018-of-00024.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00018-of-00024.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00020-of-00024.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00020-of-00024.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00020-of-00024.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00020-of-00024.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00020-of-00024.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00019-of-00024.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00019-of-00024.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00019-of-00024.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00019-of-00024.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00021-of-00024.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00021-of-00024.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00021-of-00024.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00021-of-00024.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00021-of-00024.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00020-of-00024.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00020-of-00024.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00020-of-00024.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00020-of-00024.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00004-of-00024.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00004-of-00024.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00004-of-00024.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00004-of-00024.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00004-of-00024.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00003-of-00024.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00003-of-00024.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00003-of-00024.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00003-of-00024.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00022-of-00024.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00022-of-00024.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00022-of-00024.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00022-of-00024.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00022-of-00024.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00021-of-00024.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00021-of-00024.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00021-of-00024.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00021-of-00024.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00024-of-00024.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00024-of-00024.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00024-of-00024.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00024-of-00024.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00024-of-00024.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00022-of-00024.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00022-of-00024.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00022-of-00024.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00022-of-00024.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00005-of-00024.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00005-of-00024.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00005-of-00024.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00005-of-00024.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00005-of-00024.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00004-of-00024.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00004-of-00024.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00004-of-00024.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00004-of-00024.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00006-of-00024.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00006-of-00024.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00006-of-00024.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00006-of-00024.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00006-of-00024.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00005-of-00024.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00005-of-00024.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00005-of-00024.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00005-of-00024.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00007-of-00024.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00007-of-00024.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00007-of-00024.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00007-of-00024.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00007-of-00024.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00006-of-00024.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00006-of-00024.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00006-of-00024.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00006-of-00024.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00008-of-00024.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00008-of-00024.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00008-of-00024.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00008-of-00024.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00008-of-00024.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00007-of-00024.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00007-of-00024.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00007-of-00024.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00007-of-00024.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00009-of-00024.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00009-of-00024.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00009-of-00024.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00009-of-00024.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00009-of-00024.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00008-of-00024.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00008-of-00024.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00008-of-00024.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00008-of-00024.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00010-of-00024.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00010-of-00024.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00010-of-00024.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00010-of-00024.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00010-of-00024.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00009-of-00024.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00009-of-00024.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00009-of-00024.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00009-of-00024.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00011-of-00024.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00011-of-00024.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00011-of-00024.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00011-of-00024.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00011-of-00024.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00010-of-00024.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00010-of-00024.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00010-of-00024.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00010-of-00024.safetensors",
+    "model.norm.weight": "model-00024-of-00024.safetensors"
   }
 }