Upload folder using huggingface_hub

README.md

@@ -1,3 +1,63 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+tags:
+- merge
+- mergekit
+- lazymergekit
+- MediaTek-Research/Breeze-7B-32k-Base-v1_0
+---
+
+# Breeze-13B-32k-Base-v1_0
+
+Breeze-13B-32k-Base-v1_0 is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
+* [MediaTek-Research/Breeze-7B-32k-Base-v1_0](https://huggingface.co/MediaTek-Research/Breeze-7B-32k-Base-v1_0)
+* [MediaTek-Research/Breeze-7B-32k-Base-v1_0](https://huggingface.co/MediaTek-Research/Breeze-7B-32k-Base-v1_0)
+* [MediaTek-Research/Breeze-7B-32k-Base-v1_0](https://huggingface.co/MediaTek-Research/Breeze-7B-32k-Base-v1_0)
+* [MediaTek-Research/Breeze-7B-32k-Base-v1_0](https://huggingface.co/MediaTek-Research/Breeze-7B-32k-Base-v1_0)
+* [MediaTek-Research/Breeze-7B-32k-Base-v1_0](https://huggingface.co/MediaTek-Research/Breeze-7B-32k-Base-v1_0)
+* [MediaTek-Research/Breeze-7B-32k-Base-v1_0](https://huggingface.co/MediaTek-Research/Breeze-7B-32k-Base-v1_0)
+* [MediaTek-Research/Breeze-7B-32k-Base-v1_0](https://huggingface.co/MediaTek-Research/Breeze-7B-32k-Base-v1_0)
+
+## 🧩 Configuration
+
+```yaml
+dtype: bfloat16
+merge_method: passthrough
+slices:
+- sources:
+  - layer_range: [0, 8]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [0, 8]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [4, 12]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [4, 12]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [8, 16]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [8, 16]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [12, 20]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [12, 20]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [16, 24]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [16, 24]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [20, 28]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [20, 28]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [24, 32]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [24, 32]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+
+```

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "<EOD>": 61873,
+  "<PAD>": 61874
+}

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "MediaTek-Research/Breeze-7B-32k-Base-v1_0",
+  "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": 56,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": 65536,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.41.2",
+  "use_cache": true,
+  "vocab_size": 62464
+}

mergekit_config.yml

@@ -0,0 +1,40 @@
+
+dtype: bfloat16
+merge_method: passthrough
+slices:
+- sources:
+  - layer_range: [0, 8]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [0, 8]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [4, 12]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [4, 12]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [8, 16]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [8, 16]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [12, 20]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [12, 20]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [16, 24]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [16, 24]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [20, 28]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [20, 28]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+- sources:
+  - layer_range: [24, 32]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+  - layer_range: [24, 32]
+    model: MediaTek-Research/Breeze-7B-32k-Base-v1_0    
+

model-00001-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da52ae431e35f3918a5929d19cccdc5010f03053313236268d022ac3235242b3
+size 511705216

model-00002-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8e831e3d45e710795d04101a37498c889b07b0bea4d48dd333ecf9822a3b5df
+size 947938552

model-00003-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2a5949db9300ec1ac36dd0d6596ff96ac8011d3e59e0d647f94287051984571
+size 905995864

model-00004-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9711b20a27967c3f4719431c3fef58eeea77276b553b906f548c497d5f4e1509
+size 989890728

model-00005-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b96966de3fcf6a612ae5c1a50f053e99f367d6b1f6ff661bfef154939610f71
+size 998296072

model-00006-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2060c6c57a1bd3dce20fd1ec0e51b4385a98327df6f9a8ac267ddf542925de4
+size 981502456

model-00007-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ded5620b118458b5284715798c76c07bd4e0f202bf92f99a56c7fdb51cfa6724
+size 922798696

model-00008-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c5144e6b0fa71fbba7b7f9a7c3112c190669c2d579c5cfa586f84f4544f17c0
+size 989890728

model-00009-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eaf3483b5335adf53add0e70dad43e341736f93abf8a6fc5d88ac3b765d40bb5
+size 989890728

model-00010-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:166b56644a28b69d8989173712fb2736b00b1fe8bd21b470b41c8a659bc9af1e
+size 989890728

model-00011-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cccf59383de7a38329fdbef89f290a86481a226b3cd021efeba287e62040971f
+size 989890728

model-00012-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7f2f507fe74fca17d8ab70f8b6f7d793c678a92bc507784ada7a1328f32ec67
+size 956353240

model-00013-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4af43abefc2ed9bb67502c28ae1b28197dc1e98c3250e8c5e03b59a62c033004
+size 989890728

model-00014-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6317f05c57fafb2d89621013814f130444f48867f7186d1191f4db1a33b97777
+size 989890728

model-00015-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f612758445228e6479f14fd5046428ee16562194101e1e58d9c95bc4d43242e
+size 989890728

model-00016-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:202dd1b69c66b90bdef320891d9e968be8fdebbdc4a26c32c343f73228f574b5
+size 989890728

model-00017-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b648c7658fc8c3d7f6446ebf618155b08bf477c54f13de80a017330b181386fe
+size 998296072

model-00018-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad339827569b554b821c9c81e07e603dc88d71c5a3df27eb316b8b19a02dc854
+size 981502456

model-00019-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d81f9adbf0bef25523c0a7242f1a1a0151d90d26201c87c5a7791b631fe5e0d
+size 922790392

model-00020-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04e46fbb21f95e905984ca39aba7d90dbbc69847ec5cb7ad6c70a5bbf8991654
+size 989890712

model-00021-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f0fe625efd4b3fc7a4e85f44992d4f51990c6dff325931dfc6e660e70e4d0ec
+size 989890720

model-00022-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86d6cdc49ed0e0c803f0799cfe9aec5cc3e684d1630c85c38f54a854ba250345
+size 989899016

model-00023-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f2672ed84edf475ded21013aac5f7cc3d856b319ebc3c1dcd66e68a2e62c800
+size 956353208

model-00024-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:337436bd2a1f2fa0eefca800aef5032578f20080229f2b4770dfa0997b9f8b8a
+size 989890720

model-00025-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9cfb8029115dde1ef030599fea58ccba63791771c20a7eb070ce1caf5882ff3
+size 989890720

model-00026-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdba4e12573e4f4b5d2d1b7fd235f49200791054355751d591d71e6671b6d859
+size 989890728

model-00027-of-00027.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73fd7abefd284a1499e5250d0e2aca100c48515f44717759cf66e9761cce2559
+size 520119904

model.safetensors.index.json

@@ -0,0 +1 @@
+{"metadata": {"mergekit_version": "0.0.4.2", "total_size": 25451962368}, "weight_map": {"lm_head.weight": "model-00001-of-00027.safetensors", "model.embed_tokens.weight": "model-00002-of-00027.safetensors", "model.layers.0.input_layernorm.weight": "model-00002-of-00027.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00002-of-00027.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00027.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00002-of-00027.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00027.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00027.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00027.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00002-of-00027.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00002-of-00027.safetensors", "model.layers.1.input_layernorm.weight": "model-00002-of-00027.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00003-of-00027.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00003-of-00027.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00003-of-00027.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00003-of-00027.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00003-of-00027.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00003-of-00027.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00003-of-00027.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00003-of-00027.safetensors", "model.layers.18.input_layernorm.weight": "model-00003-of-00027.safetensors", "model.layers.14.input_layernorm.weight": "model-00003-of-00027.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00003-of-00027.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00003-of-00027.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00003-of-00027.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00003-of-00027.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00004-of-00027.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00004-of-00027.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00004-of-00027.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00027.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00027.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00027.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00027.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00027.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00027.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00027.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00027.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00027.safetensors", "model.layers.19.input_layernorm.weight": "model-00004-of-00027.safetensors", "model.layers.15.input_layernorm.weight": "model-00004-of-00027.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00004-of-00027.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00004-of-00027.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00004-of-00027.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00027.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00004-of-00027.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00005-of-00027.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00027.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00005-of-00027.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00027.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00005-of-00027.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00027.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00005-of-00027.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00027.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00005-of-00027.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00027.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00005-of-00027.safetensors", "model.layers.24.input_layernorm.weight": "model-00005-of-00027.safetensors", "model.layers.20.input_layernorm.weight": "model-00005-of-00027.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00005-of-00027.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00005-of-00027.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00005-of-00027.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00027.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00005-of-00027.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00005-of-00027.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00005-of-00027.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00027.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00005-of-00027.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00006-of-00027.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00027.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00006-of-00027.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00027.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00006-of-00027.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00027.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00006-of-00027.safetensors", "model.layers.25.input_layernorm.weight": "model-00006-of-00027.safetensors", "model.layers.21.input_layernorm.weight": "model-00006-of-00027.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00006-of-00027.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00006-of-00027.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00027.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00006-of-00027.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00006-of-00027.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00006-of-00027.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00027.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00006-of-00027.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00027.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00006-of-00027.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00027.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00006-of-00027.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00027.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00007-of-00027.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00007-of-00027.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00007-of-00027.safetensors", "model.layers.26.input_layernorm.weight": "model-00007-of-00027.safetensors", "model.layers.22.input_layernorm.weight": "model-00007-of-00027.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00007-of-00027.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00007-of-00027.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00007-of-00027.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00007-of-00027.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00007-of-00027.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00007-of-00027.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00007-of-00027.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00007-of-00027.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00007-of-00027.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00007-of-00027.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00007-of-00027.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00007-of-00027.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00007-of-00027.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00007-of-00027.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00007-of-00027.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00007-of-00027.safetensors", "model.layers.27.input_layernorm.weight": "model-00007-of-00027.safetensors", "model.layers.23.input_layernorm.weight": "model-00007-of-00027.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00008-of-00027.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00008-of-00027.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00008-of-00027.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00008-of-00027.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00008-of-00027.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00008-of-00027.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00008-of-00027.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00008-of-00027.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00008-of-00027.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00008-of-00027.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00008-of-00027.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00008-of-00027.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00008-of-00027.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00008-of-00027.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00008-of-00027.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00008-of-00027.safetensors", "model.layers.32.input_layernorm.weight": "model-00008-of-00027.safetensors", "model.layers.28.input_layernorm.weight": "model-00008-of-00027.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00008-of-00027.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00009-of-00027.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00009-of-00027.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00009-of-00027.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00009-of-00027.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00009-of-00027.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00009-of-00027.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00009-of-00027.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00009-of-00027.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00009-of-00027.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00009-of-00027.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00009-of-00027.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00009-of-00027.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00009-of-00027.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00009-of-00027.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00009-of-00027.safetensors", "model.layers.33.input_layernorm.weight": "model-00009-of-00027.safetensors", "model.layers.29.input_layernorm.weight": "model-00009-of-00027.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00009-of-00027.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00009-of-00027.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00010-of-00027.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00010-of-00027.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00010-of-00027.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00010-of-00027.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00010-of-00027.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00010-of-00027.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00010-of-00027.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00010-of-00027.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00010-of-00027.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00010-of-00027.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00010-of-00027.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00010-of-00027.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00010-of-00027.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00010-of-00027.safetensors", "model.layers.34.input_layernorm.weight": "model-00010-of-00027.safetensors", "model.layers.30.input_layernorm.weight": "model-00010-of-00027.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00010-of-00027.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00010-of-00027.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00010-of-00027.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00011-of-00027.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00011-of-00027.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00011-of-00027.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00011-of-00027.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00011-of-00027.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00011-of-00027.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00011-of-00027.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00011-of-00027.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00011-of-00027.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00011-of-00027.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00011-of-00027.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00011-of-00027.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00011-of-00027.safetensors", "model.layers.35.input_layernorm.weight": "model-00011-of-00027.safetensors", "model.layers.31.input_layernorm.weight": "model-00011-of-00027.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00011-of-00027.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00011-of-00027.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00011-of-00027.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00011-of-00027.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00012-of-00027.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00012-of-00027.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00012-of-00027.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00012-of-00027.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00012-of-00027.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00012-of-00027.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00012-of-00027.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00012-of-00027.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00012-of-00027.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00012-of-00027.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00012-of-00027.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00012-of-00027.safetensors", "model.layers.2.input_layernorm.weight": "model-00012-of-00027.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00012-of-00027.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00012-of-00027.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00012-of-00027.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00012-of-00027.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00012-of-00027.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00012-of-00027.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00012-of-00027.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00012-of-00027.safetensors", "model.layers.40.input_layernorm.weight": "model-00012-of-00027.safetensors", "model.layers.36.input_layernorm.weight": "model-00012-of-00027.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00012-of-00027.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00013-of-00027.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00013-of-00027.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00013-of-00027.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00013-of-00027.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00013-of-00027.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00013-of-00027.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00013-of-00027.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00013-of-00027.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00013-of-00027.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00013-of-00027.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00013-of-00027.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00013-of-00027.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00013-of-00027.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00013-of-00027.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00013-of-00027.safetensors", "model.layers.41.input_layernorm.weight": "model-00013-of-00027.safetensors", "model.layers.37.input_layernorm.weight": "model-00013-of-00027.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00013-of-00027.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00013-of-00027.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00014-of-00027.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00014-of-00027.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00014-of-00027.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00014-of-00027.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00014-of-00027.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00014-of-00027.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00014-of-00027.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00014-of-00027.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00014-of-00027.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00014-of-00027.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00014-of-00027.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00014-of-00027.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00014-of-00027.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00014-of-00027.safetensors", "model.layers.42.input_layernorm.weight": "model-00014-of-00027.safetensors", "model.layers.38.input_layernorm.weight": "model-00014-of-00027.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00014-of-00027.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00014-of-00027.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00014-of-00027.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00015-of-00027.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00015-of-00027.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00015-of-00027.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00015-of-00027.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00015-of-00027.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00015-of-00027.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00015-of-00027.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00015-of-00027.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00015-of-00027.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00015-of-00027.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00015-of-00027.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00015-of-00027.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00015-of-00027.safetensors", "model.layers.43.input_layernorm.weight": "model-00015-of-00027.safetensors", "model.layers.39.input_layernorm.weight": "model-00015-of-00027.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00015-of-00027.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00015-of-00027.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00015-of-00027.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00015-of-00027.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00016-of-00027.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00016-of-00027.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00016-of-00027.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00016-of-00027.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00016-of-00027.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00016-of-00027.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00016-of-00027.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00016-of-00027.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00016-of-00027.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00016-of-00027.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00016-of-00027.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00016-of-00027.safetensors", "model.layers.48.input_layernorm.weight": "model-00016-of-00027.safetensors", "model.layers.44.input_layernorm.weight": "model-00016-of-00027.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00016-of-00027.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00016-of-00027.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00016-of-00027.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00016-of-00027.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00016-of-00027.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00017-of-00027.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00017-of-00027.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00017-of-00027.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00017-of-00027.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00017-of-00027.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00017-of-00027.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00017-of-00027.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00017-of-00027.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00017-of-00027.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00017-of-00027.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00017-of-00027.safetensors", "model.layers.49.input_layernorm.weight": "model-00017-of-00027.safetensors", "model.layers.45.input_layernorm.weight": "model-00017-of-00027.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00017-of-00027.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00017-of-00027.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00017-of-00027.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00017-of-00027.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00017-of-00027.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00017-of-00027.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00017-of-00027.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00017-of-00027.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00017-of-00027.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00018-of-00027.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00018-of-00027.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00018-of-00027.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00018-of-00027.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00018-of-00027.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00018-of-00027.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00018-of-00027.safetensors", "model.layers.50.input_layernorm.weight": "model-00018-of-00027.safetensors", "model.layers.46.input_layernorm.weight": "model-00018-of-00027.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00018-of-00027.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00018-of-00027.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00018-of-00027.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00018-of-00027.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00018-of-00027.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00018-of-00027.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00018-of-00027.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00018-of-00027.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00018-of-00027.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00018-of-00027.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00018-of-00027.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00018-of-00027.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00018-of-00027.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00019-of-00027.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00019-of-00027.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00019-of-00027.safetensors", "model.layers.51.input_layernorm.weight": "model-00019-of-00027.safetensors", "model.layers.47.input_layernorm.weight": "model-00019-of-00027.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00019-of-00027.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00019-of-00027.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00019-of-00027.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00019-of-00027.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00019-of-00027.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00019-of-00027.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00019-of-00027.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00019-of-00027.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00019-of-00027.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00019-of-00027.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00019-of-00027.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00019-of-00027.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00019-of-00027.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00019-of-00027.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00019-of-00027.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00019-of-00027.safetensors", "model.layers.52.input_layernorm.weight": "model-00019-of-00027.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00020-of-00027.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00020-of-00027.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00020-of-00027.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00020-of-00027.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00020-of-00027.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00020-of-00027.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00020-of-00027.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00020-of-00027.safetensors", "model.layers.53.input_layernorm.weight": "model-00020-of-00027.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00020-of-00027.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00020-of-00027.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00020-of-00027.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00020-of-00027.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00020-of-00027.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00020-of-00027.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00020-of-00027.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00020-of-00027.safetensors", "model.layers.3.input_layernorm.weight": "model-00020-of-00027.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00020-of-00027.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00021-of-00027.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00021-of-00027.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00021-of-00027.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00021-of-00027.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00021-of-00027.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00021-of-00027.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00021-of-00027.safetensors", "model.layers.54.input_layernorm.weight": "model-00021-of-00027.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00021-of-00027.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00021-of-00027.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00021-of-00027.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00021-of-00027.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00021-of-00027.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00021-of-00027.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00021-of-00027.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00021-of-00027.safetensors", "model.layers.55.input_layernorm.weight": "model-00021-of-00027.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00021-of-00027.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00021-of-00027.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00022-of-00027.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00022-of-00027.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00022-of-00027.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00022-of-00027.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00022-of-00027.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00022-of-00027.safetensors", "model.layers.8.input_layernorm.weight": "model-00022-of-00027.safetensors", "model.layers.4.input_layernorm.weight": "model-00022-of-00027.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00022-of-00027.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00022-of-00027.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00022-of-00027.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00022-of-00027.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00022-of-00027.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00022-of-00027.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00022-of-00027.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00022-of-00027.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00022-of-00027.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00022-of-00027.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00022-of-00027.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00022-of-00027.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00023-of-00027.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00023-of-00027.safetensors", "model.layers.8.self_attn.v_proj.weight": "model-00023-of-00027.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00023-of-00027.safetensors", "model.layers.9.input_layernorm.weight": "model-00023-of-00027.safetensors", "model.layers.5.input_layernorm.weight": "model-00023-of-00027.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00023-of-00027.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00023-of-00027.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00023-of-00027.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00023-of-00027.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00023-of-00027.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00023-of-00027.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00023-of-00027.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00023-of-00027.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00023-of-00027.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00023-of-00027.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00023-of-00027.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00023-of-00027.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00023-of-00027.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00023-of-00027.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00023-of-00027.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00023-of-00027.safetensors", "model.layers.10.input_layernorm.weight": "model-00023-of-00027.safetensors", "model.layers.6.input_layernorm.weight": "model-00023-of-00027.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00024-of-00027.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00024-of-00027.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00024-of-00027.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00024-of-00027.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00024-of-00027.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00024-of-00027.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00024-of-00027.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00024-of-00027.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00024-of-00027.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00024-of-00027.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00024-of-00027.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00024-of-00027.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00024-of-00027.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00024-of-00027.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00024-of-00027.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00024-of-00027.safetensors", "model.layers.11.input_layernorm.weight": "model-00024-of-00027.safetensors", "model.layers.7.input_layernorm.weight": "model-00024-of-00027.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00024-of-00027.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00025-of-00027.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00025-of-00027.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00025-of-00027.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00025-of-00027.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00025-of-00027.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00025-of-00027.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00025-of-00027.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00025-of-00027.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00025-of-00027.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00025-of-00027.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00025-of-00027.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00025-of-00027.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00025-of-00027.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00025-of-00027.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00025-of-00027.safetensors", "model.layers.16.input_layernorm.weight": "model-00025-of-00027.safetensors", "model.layers.12.input_layernorm.weight": "model-00025-of-00027.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00025-of-00027.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00025-of-00027.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00026-of-00027.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00026-of-00027.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00026-of-00027.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00026-of-00027.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00026-of-00027.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00026-of-00027.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00026-of-00027.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00026-of-00027.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00026-of-00027.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00026-of-00027.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00026-of-00027.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00026-of-00027.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00026-of-00027.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00026-of-00027.safetensors", "model.layers.17.input_layernorm.weight": "model-00026-of-00027.safetensors", "model.layers.13.input_layernorm.weight": "model-00026-of-00027.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00026-of-00027.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00026-of-00027.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00026-of-00027.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00027-of-00027.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00027-of-00027.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00027-of-00027.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00027-of-00027.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00027-of-00027.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00027-of-00027.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00027-of-00027.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00027-of-00027.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00027-of-00027.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00027-of-00027.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00027-of-00027.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00027-of-00027.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00027-of-00027.safetensors", "model.norm.weight": "model-00027-of-00027.safetensors"}}

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "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
+  },
+  "pad_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 sha256:9298e56c094f0d30431b0e52ad53287f0cadc99ac8ca17cc2144b0eb4753f130
+size 911034

tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "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
+    },
+    "61873": {
+      "content": "<EOD>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "61874": {
+      "content": "<PAD>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}