Upload folder using huggingface_hub

chat_template.jinja

@@ -0,0 +1,24 @@
+{%- if messages[0]['role'] == 'system' %}
+    {%- set system_message = messages[0]['content'] %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+
+{{- bos_token }}
+{%- for message in loop_messages %}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}
+        {{- raise_exception('After the optional system message, conversation roles must alternate user/assistant/user/assistant/...') }}
+    {%- endif %}
+    {%- if message['role'] == 'user' %}
+        {%- if loop.first and system_message is defined %}
+            {{- ' [INST] ' + system_message + '\n\n' + message['content'] + ' [/INST]' }}
+        {%- else %}
+            {{- ' [INST] ' + message['content'] + ' [/INST]' }}
+        {%- endif %}
+    {%- elif message['role'] == 'assistant' %}
+        {{- ' ' + message['content'] + eos_token}}
+    {%- 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 +1,26 @@
+{
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "head_dim": null,
+  "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,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.1",
+  "use_cache": true,
+  "vocab_size": 32000
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.57.1"
+}

model-00001-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61109e24b7de62d4e176e5c5e4963f36eb3f8aacb28f169e9cb174f5273d7c88
+size 926941888

model-00002-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1439328e656bafa4782d816870bf0b37889c47b0374c48a3dc6ac883d131b48c
+size 872449040

model-00003-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b63c1053a1cc79981b6027e00746c56395562364ac544d7e0013357f7649fd3
+size 872449040

model-00004-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f659ecab5d6a0f8ce3d2afcfcf9f837060df2e4a2a8ece92285b8ad507c6ffb
+size 872449040

model-00005-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca62869fc2a989decda5c717b8eb6996501bc7f3b0faf2e4d39249ca40fe51fc
+size 872449040

model-00006-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:009d2ca90012a571d171fd1698129466f492f46820e7bd00ebafe0c6e6022b74
+size 872449040

model-00007-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98ed9db086d2e0a35074bb20cd6e1dc3516f33822320ae8b1d88a5a28f645306
+size 872449040

model-00008-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e68df08ebbcee6a61adf69435674e7c50d2278d57dfbe806740e6578218136b
+size 872449040

model-00009-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9d90baa439304c9237ce672cc5a81d834f8162349288b6bbb3c1563d0294bf0
+size 872449040

model-00010-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63e0fdcb9970d5544bd02d8fadebbf45fe59b88135069f8588cc498489d30830
+size 872449040

model-00011-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9af9734397d827e9651455c3c6a42b73998e0a59245dc344007f984594dfa6b7
+size 872449056

model-00012-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dcbd25af4ca698971caf380651acb11ca5ba6bc4ad6e61ced25a52932d6105c
+size 872449048

model-00013-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1437540797e2a7a57f0fac5d0e34203763cacf3e11dc4098b102659e8f58e15
+size 872449048

model-00014-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e0fd584930499e1a986d875500af29cce636862d01306d5b1d81621908d3c5b
+size 872449048

model-00015-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37791d91d7b16aa376f211630d8d87a933db7214491e68b16b2385553583fce8
+size 872449048

model-00016-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62adcc070119cda37fe814187cf01f1b770ab34727d0462e5d55509d36e879cd
+size 872449048

model-00017-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87bf3ee77f479161457989c604aebe737b57303e6d4979afdfc0eecdc4d068b6
+size 872449048

model-00018-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4c6677e4f8d2f16129dcbdfee349b9e2053b8c24a7c0fa2c2f34e2ee06f86c5
+size 872449048

model-00019-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c6bb8a8cbabf39a30acdec53d0b712f752e6c745130efbba5f3e15df33a57c5
+size 872449048

model-00020-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d2a3361ad7233beb535331d87a708c68925da7f44d143105c05ffe8317ec67c
+size 872449048

model-00021-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3773ea7efb9da888aa71df372d616604ca4941fa728ad3add484bb794d5169c5
+size 872449048

model-00022-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:500ca0423d121cd0ffab0b40087b1f0672f0feb7d3d7f0ff4d62f257ec1b3da3
+size 872449048

model-00023-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c45776ebd5cf547818c995db34684c668d7c7887efe8ecfa519f19c7f6daaa9f
+size 872449048

model-00024-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4065b0279996c3e81dc86bbe4415a5a50398ac66eb0994500f0bdc21f66b4f6
+size 872449048

model-00025-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f737d55b3d0049476dd9a7c192f492e2558ecaf8102d8f9784070648d02723f
+size 872449048

model-00026-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e7ae11fc52e7b3f4fdb4a12960c1972b1d8d3104b7aedfda68a6de780890699
+size 872449048

model-00027-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dc68284c87ceb1e94b5648cd542d6da22d782ed9e23888c404f4211708dd806
+size 872449048

model-00028-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31c29e5683beda682b363db1b91eb45a0881a882ee70ab4f203291d6139ec61c
+size 872449048

model-00029-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2097247e67b206afa29f1119bf1d6015224c852adf03ebbd364c593953b8f3b0
+size 872449048

model-00030-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e12dddf718b9d0997db7d671b5d7f229ca878f842b632d550d63276a033bbf5
+size 872449048

model-00031-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d81a91de75956265f680ce8d7612c1fc66a88aab88a29ff1c115f3839db49fb
+size 872449048

model-00032-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:845d08a4972edb84ddff7545e60d837883d133ea9f6e208e17935c824839a1b8
+size 872449048

model-00033-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3347c556b9e97f89a43b582ababad76fd588343a4a761822715225fff68cb68
+size 994099864

model.safetensors.index.json

@@ -0,0 +1,299 @@
+{
+  "metadata": {
+    "total_parameters": 7241732096,
+    "total_size": 28966928384
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00033-of-00033.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00033.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00002-of-00033.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00002-of-00033.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00033.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00002-of-00033.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00033.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00033.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00033.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00033.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00033.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00003-of-00033.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00003-of-00033.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00033.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00003-of-00033.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00003-of-00033.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00033.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00033.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00033.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00033.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00012-of-00033.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00012-of-00033.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00011-of-00033.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00012-of-00033.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00012-of-00033.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00011-of-00033.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00011-of-00033.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00011-of-00033.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00011-of-00033.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00013-of-00033.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00013-of-00033.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00012-of-00033.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00013-of-00033.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00013-of-00033.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00012-of-00033.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00012-of-00033.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00012-of-00033.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00012-of-00033.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00014-of-00033.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00014-of-00033.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00013-of-00033.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00014-of-00033.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00014-of-00033.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00013-of-00033.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00013-of-00033.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00013-of-00033.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00013-of-00033.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00015-of-00033.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00015-of-00033.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00014-of-00033.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00015-of-00033.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00015-of-00033.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00014-of-00033.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00014-of-00033.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00014-of-00033.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00014-of-00033.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00016-of-00033.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00016-of-00033.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00015-of-00033.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00016-of-00033.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00016-of-00033.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00015-of-00033.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00015-of-00033.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00015-of-00033.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00015-of-00033.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00017-of-00033.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00017-of-00033.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00016-of-00033.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00017-of-00033.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00017-of-00033.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00016-of-00033.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00016-of-00033.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00016-of-00033.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00016-of-00033.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00018-of-00033.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00018-of-00033.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00017-of-00033.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00018-of-00033.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00018-of-00033.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00017-of-00033.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00017-of-00033.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00017-of-00033.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00017-of-00033.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00019-of-00033.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00019-of-00033.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00018-of-00033.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00019-of-00033.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00019-of-00033.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00018-of-00033.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00018-of-00033.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00018-of-00033.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00018-of-00033.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00020-of-00033.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00020-of-00033.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00019-of-00033.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00020-of-00033.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00020-of-00033.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00019-of-00033.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00019-of-00033.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00019-of-00033.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00019-of-00033.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00021-of-00033.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00021-of-00033.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00020-of-00033.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00021-of-00033.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00021-of-00033.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00020-of-00033.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00020-of-00033.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00020-of-00033.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00020-of-00033.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00004-of-00033.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00004-of-00033.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00003-of-00033.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00004-of-00033.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00004-of-00033.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00003-of-00033.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00003-of-00033.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00003-of-00033.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00003-of-00033.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00022-of-00033.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00022-of-00033.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00021-of-00033.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00022-of-00033.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00022-of-00033.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00021-of-00033.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00021-of-00033.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00021-of-00033.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00021-of-00033.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00023-of-00033.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00023-of-00033.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00022-of-00033.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00023-of-00033.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00023-of-00033.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00022-of-00033.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00022-of-00033.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00022-of-00033.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00022-of-00033.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00024-of-00033.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00024-of-00033.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00023-of-00033.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00024-of-00033.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00024-of-00033.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00023-of-00033.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00023-of-00033.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00023-of-00033.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00023-of-00033.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00025-of-00033.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00025-of-00033.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00024-of-00033.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00025-of-00033.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00025-of-00033.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00024-of-00033.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00024-of-00033.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00024-of-00033.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00024-of-00033.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00026-of-00033.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00026-of-00033.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00025-of-00033.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00026-of-00033.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00026-of-00033.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00025-of-00033.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00025-of-00033.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00025-of-00033.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00025-of-00033.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00027-of-00033.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00027-of-00033.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00026-of-00033.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00027-of-00033.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00027-of-00033.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00026-of-00033.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00026-of-00033.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00026-of-00033.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00026-of-00033.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00028-of-00033.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00028-of-00033.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00027-of-00033.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00028-of-00033.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00028-of-00033.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00027-of-00033.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00027-of-00033.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00027-of-00033.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00027-of-00033.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00029-of-00033.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00029-of-00033.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00028-of-00033.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00029-of-00033.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00029-of-00033.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00028-of-00033.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00028-of-00033.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00028-of-00033.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00028-of-00033.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00030-of-00033.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00030-of-00033.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00029-of-00033.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00030-of-00033.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00030-of-00033.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00029-of-00033.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00029-of-00033.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00029-of-00033.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00029-of-00033.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00031-of-00033.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00031-of-00033.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00030-of-00033.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00031-of-00033.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00031-of-00033.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00030-of-00033.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00030-of-00033.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00030-of-00033.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00030-of-00033.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00005-of-00033.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00005-of-00033.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00004-of-00033.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00005-of-00033.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00005-of-00033.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00004-of-00033.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00004-of-00033.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00004-of-00033.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00004-of-00033.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00032-of-00033.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00032-of-00033.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00031-of-00033.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00032-of-00033.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00032-of-00033.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00031-of-00033.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00031-of-00033.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00031-of-00033.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00031-of-00033.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00033-of-00033.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00033-of-00033.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00032-of-00033.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00033-of-00033.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00033-of-00033.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00032-of-00033.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00032-of-00033.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00032-of-00033.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00032-of-00033.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00006-of-00033.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00006-of-00033.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00005-of-00033.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00006-of-00033.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00006-of-00033.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00005-of-00033.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00005-of-00033.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00005-of-00033.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00005-of-00033.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00007-of-00033.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00007-of-00033.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00006-of-00033.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00007-of-00033.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00007-of-00033.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00006-of-00033.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00006-of-00033.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00006-of-00033.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00006-of-00033.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00008-of-00033.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00008-of-00033.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00007-of-00033.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00008-of-00033.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00008-of-00033.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00007-of-00033.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00007-of-00033.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00007-of-00033.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00007-of-00033.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00009-of-00033.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00009-of-00033.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00008-of-00033.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00009-of-00033.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00009-of-00033.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00008-of-00033.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00008-of-00033.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00008-of-00033.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00008-of-00033.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00010-of-00033.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00010-of-00033.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00009-of-00033.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00010-of-00033.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00010-of-00033.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00009-of-00033.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00009-of-00033.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00009-of-00033.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00009-of-00033.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00011-of-00033.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00011-of-00033.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00010-of-00033.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00011-of-00033.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00011-of-00033.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00010-of-00033.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00010-of-00033.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00010-of-00033.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00010-of-00033.safetensors",
+    "model.norm.weight": "model-00033-of-00033.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}