Upload MistralForCausalLM

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "./models_merged",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "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,
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.2",
+  "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": "4.36.2"
+}

model-00001-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a0ff08e9613ed1ec5afeef167b72f57086732d95a510bbe4d1305cf3db89139
+size 926941888

model-00002-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6ce9ae7ac3d1491b12abdb781fd835e3ef46a37489b6e64f8e531438603a6ed
+size 872449040

model-00003-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c46d5bee7cc201e8713fc16fee233b19b0382a68a4938de4efe1f7b04d5caa31
+size 872449040

model-00004-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2692b0a2d87a7c307c3f965785b1bc4a820f692f1ecf47cfad20fabb800183ae
+size 872449040

model-00005-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8064d1c4178439dcbb5cfeb249eeac72ea222cbf10cab817790ad6cb698355e2
+size 872449040

model-00006-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ba6782e57fe114330ad4b1ee493e72159e18f5b4c68be5b5349d6603ad63440
+size 872449040

model-00007-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2115b212ae135595e038f3a892a14cff34f561e796bc2f64c4c9e44048db765f
+size 872449040

model-00008-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:808c4e5a024af3194debcf26f11c648054a35c9e9b970faaa6f282cad85ebcb1
+size 872449040

model-00009-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fdd0c43a0513b704778816ef22fcc85cfe9766b116993b4998a979477fb74f7
+size 872449040

model-00010-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14077c59af0c38eb193bbf2310001eaa993ef7dfea65cce9ba20796507f92a4c
+size 872449040

model-00011-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b27e1c642959e4c86b3d5c092f0c327450ab55d119ad598803af14e97570c47
+size 872449056

model-00012-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd029b1a73181bc1b0c13678cddf37ff5664874b1f274cf45af6fec8f79ebd70
+size 872449048

model-00013-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:281f453ebffd9c3d1caf519880e0596bd2aa31fa4e79a8f36789572e70e4c49d
+size 872449048

model-00014-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8266fba8e3d7b65472b19d46152c22eaaef731eb78599a90532054ebcd8e8c91
+size 872449048

model-00015-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86f0d91be61d7128ed840464ec6f1d9ef64e06039648a960efce4b64a2134dd4
+size 872449048

model-00016-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43feebd6e398acf23c97c475b908f2b40189b53b3e68b87ebd2c8123beb642ba
+size 872449048

model-00017-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e9d5efd99211e9c727683f1359ed1d9f3fe0574f8b611c20a40b39caa000dfd
+size 872449048

model-00018-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c016b17a6f9d355d73b2eb4eb42ca11294fd299b4639be65cb941437b9e93ee4
+size 872449048

model-00019-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe92d5fe7f0b13491967e003323d410a87739c24991527fa9283d1f85af7f952
+size 872449048

model-00020-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68d92e251c1d70611595ac964847ad6effe14067107473fc086484e93328db6e
+size 872449048

model-00021-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcf666d585ccd9f7a90d532f3f80e554229fcd0146a637f627371a3616964448
+size 872449048

model-00022-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d18a7f6fd064518803c3c81b73bc613c7e54515eb7ebad57e706ac1ba186aae0
+size 872449048

model-00023-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7c3a12efb8e34b25d07b00237f6bc0f84c6f331491da4c5a178e3e82e29b56f
+size 872449048

model-00024-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2191df8ffc125cfc6b9a374ba218f259e0847413d9c15c4e2572f8b75749090a
+size 872449048

model-00025-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90f586c4a260d5ca6b263cf9c804d62d0e571339945f3ba9587b02d2efe0cffb
+size 872449048

model-00026-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ed978115ac9805f7864a75ba6aa5e9cb99ddbdb214b34a55fcbb468fbb1e85e
+size 872449048

model-00027-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab44011a5d25bcece4372b2d3b45ff7acb53f58bf79db36bfa6235105e0ce956
+size 872449048

model-00028-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbe264030d75e8eeba80625cbcaa4f89b16e881df2fd84b85f9fb5c894b4013e
+size 872449048

model-00029-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ec6e4cfeb5c3e2de64b81c0da39c3e5c8261d94f2f8c49fd73e3254a9094f21
+size 872449048

model-00030-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22272ad080f92419375bf08cb1081048a08415faa535aab60e63479f6472e8ed
+size 872449048

model-00031-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62af133695830a06317c998af1369a5560a6f252e7f75310a1f9ba43656039de
+size 872449048

model-00032-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73962c6ca344975893d4a63194bf58291aac7266447e550040c66f2f0114ebca
+size 872449048

model-00033-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b0f87fed59fad38bbe776080f7f181c64426630acb5c5fb9fa6e3aa4889195f
+size 994099864

model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "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"
+  }
+}