Upload MistralForCausalLM

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "/data/llm_leaderboard/model_results/20240104_init-test",
+  "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 sha256:0d08209cb6c5f40efc4d1e57f5a98879115ae2a007f29befa7a260dfede5250d
+size 926941888

model-00002-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0036c5651c6ce7700d9ca871e5b290eab901ce3c722dc1d28c6bbe322072bb6e
+size 872449040

model-00003-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8095c6bbbf452c44172dc2469abe56e6634993fca5cb05f1572851cd9be90e7a
+size 872449040

model-00004-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fc3081e0e907f307fc324f20aea907be7bfd9e1ea8c209c0b3982feeef847e6
+size 872449040

model-00005-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:914cd8f76851dda54fff675738e698092153c455be0a9940cee8d7f5ea82187e
+size 872449040

model-00006-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c06b8096296a17c8c5cf7daab725bf99ae1cf267148e40206b951ef1af8124c1
+size 872449040

model-00007-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76ea1034b00511da1db657738a1a6b29741d0e9cb1267c788aa218fa27e7aa17
+size 872449040

model-00008-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:529640e02e95f30fb88ab5f8ea68123b5114d98c9ec27cbf6f76978c69f6d505
+size 872449040

model-00009-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a0be3456d53574d11567e7bb26eb0f489e49cdf5c008c6c768d81c318ffa3dd
+size 872449040

model-00010-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52daa31de588ae47ab48934d00c763b795c8bde9aa9d1521bcd242fd0c8559ad
+size 872449040

model-00011-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e30cc911bec98468f4ff8a13b22eb91a52847e0a264e6a4961edf9c0e7aaf805
+size 872449056

model-00012-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d94704ee5e5d6962e0d64556ebc913bf0ae95ff596b95ea17bf2a8df03e35e9e
+size 872449048

model-00013-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0162a4ca100c5a31eb714b75dcbcc68548a6990a97f5853e62737dbee65065ed
+size 872449048

model-00014-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31406308966fe7952c07c8169417a97f83ff872d0dd614c77b0eca80d7dee0a2
+size 872449048

model-00015-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:836a34b341bc7cd88ba2f2fa64629e28538f58a03e062a1bc626c461bf4d85aa
+size 872449048

model-00016-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e3b777482c1685630141f57ac16341ef070c0f3ffa6c1dbce47a1ebd640f1a2
+size 872449048

model-00017-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:106f0113dc5b9f23595b77e39667664b098886710e375a12e7271fd87ccdd5d2
+size 872449048

model-00018-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f55a3eb3ab2d9b8010adeeaa6b7b91d9a322840a7bded88fa28a68af07a26652
+size 872449048

model-00019-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5237685f5763d037bdcdacfdec01f04e5ea6056b59778d08795fe43727321500
+size 872449048

model-00020-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ab8b5d67d2702f62c4b24105f659811b222d87740f9a3f04f445837d8a2ba3e
+size 872449048

model-00021-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4b3a9f3154702b63ec86bd8a4d64c41b2e5b06e1887e6197c72d7a17a86d74a
+size 872449048

model-00022-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa054adbe1d6d81ff55c9cfb9f44a0fd13752ae4805023dbbb69e9c09fcee9b6
+size 872449048

model-00023-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed46122ff628000dc715474d6b78a84d84ad47f34f34bc8924c17c63d271a879
+size 872449048

model-00024-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c480a71003cb84220f2ba16a87917b52a240bf8a4fad013a4de80535dfbe4de4
+size 872449048

model-00025-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd17ffd97cbdfbf662b7829f6dc2a5791e298cc87a55a15a70deaf2370fba550
+size 872449048

model-00026-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec766420af0fc220ed4cf11d822adc10d5e337d9e0310e5c71b6173d17f9f7b0
+size 872449048

model-00027-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fa8f45976d2277544843111a93d3513a067ddbc0e6dd97df083921a550cf351
+size 872449048

model-00028-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a2b43956d283140116d1f01f0b78da460ef9f7c8f14c007ba0547fa033a956f
+size 872449048

model-00029-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40f11a5c5704a08c2237ad98316b96c7b811a1d4fb4e7937c9a01bdb50f25ffa
+size 872449048

model-00030-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea92ccf688acc45a43a3748b16a6cd1d5abc61670f598b80813797e0937d9174
+size 872449048

model-00031-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b68f28983c90512d6dc7b8ac871d2932ee926e7765b7d02e537ac3429f3cef6
+size 872449048

model-00032-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6dfecfae434ee034ae48577df9ac5e4e0629fb1859fdc9b4d2f3919fa551eca1
+size 872449048

model-00033-of-00033.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8dd011bf934aa8a11736d483404b66572fc08c00dce750b515718e2f265ddd5
+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"
+  }
+}