Dracones commited on
Commit
1793deb
1 Parent(s): d013cc2

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: []
3
+ library_name: transformers
4
+ tags:
5
+ - mergekit
6
+ - merge
7
+ ---
8
+
9
+ <div style="width: auto; margin-left: auto; margin-right: auto">
10
+ <img src="https://i.imgur.com/Tn9MBg6.png" alt="MidnightMiqu" style="width: 100%; min-width: 400px; display: block; margin: auto;">
11
+ </div>
12
+
13
+ # Midnight-Miqu-103B-v1.0 - EXL2 4.5bpw
14
+
15
+ This is a 4.5bpw EXL2 quant of [sophosympatheia/Midnight-Miqu-103B-v1.0](https://huggingface.co/sophosympatheia/Midnight-Miqu-103B-v1.0)
16
+
17
+ Details about the model and the merge info can be found at the above mode page.
18
+
19
+
20
+ ## Quant Details
21
+
22
+ This is the script used for quantization.
23
+
24
+ ```bash
25
+ #!/bin/bash
26
+
27
+ # Activate the conda environment
28
+ source ~/miniconda3/etc/profile.d/conda.sh
29
+ conda activate exllamav2
30
+
31
+ # Define variables
32
+ MODEL_DIR="models/sophosympatheia_Midnight-Miqu-103B-v1.0"
33
+ OUTPUT_DIR="exl2_midnight103b"
34
+ MEASUREMENT_FILE="measurements/midnight103b.json"
35
+
36
+ BIT_PRECISION=4.5
37
+ CONVERTED_FOLDER="models/Midnight-Miqu-103B_exl2_4.5bpw"
38
+
39
+ # Create directories
40
+ mkdir $OUTPUT_DIR
41
+ mkdir $CONVERTED_FOLDER
42
+
43
+ # Run conversion commands
44
+ python convert.py -i $MODEL_DIR -o $OUTPUT_DIR -nr -om $MEASUREMENT_FILE
45
+ python convert.py -i $MODEL_DIR -o $OUTPUT_DIR -nr -m $MEASUREMENT_FILE -b $BIT_PRECISION -cf $CONVERTED_FOLDER
46
+ ```
47
+
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "midnight-miqu-103b-v1.0",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 8192,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 28672,
14
+ "max_position_embeddings": 32764,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 64,
17
+ "num_hidden_layers": 120,
18
+ "num_key_value_heads": 8,
19
+ "pad_token_id": 0,
20
+ "pretraining_tp": 1,
21
+ "rms_norm_eps": 1e-05,
22
+ "rope_scaling": null,
23
+ "rope_theta": 1000000,
24
+ "tie_word_embeddings": false,
25
+ "torch_dtype": "float16",
26
+ "transformers_version": "4.36.2",
27
+ "use_cache": true,
28
+ "vocab_size": 32000
29
+ }
huggingface-metadata.txt ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ url: https://huggingface.co/sophosympatheia/Midnight-Miqu-103B-v1.0
2
+ branch: main
3
+ download date: 2024-03-04 13:53:26
4
+ sha256sum:
5
+ 3f9d7074f388e45b1c225c0663880ae541eed8b12e387daa53111a7349042470 model-00001-of-00021.safetensors
6
+ 62a672effb2b7a324329e87493e63635e830379277bb3cc688ce63b606fe9c6a model-00002-of-00021.safetensors
7
+ 6e1a937be61a99efd437ef1794ebc06ea368ae5ceda6c2a1d68165383126394c model-00003-of-00021.safetensors
8
+ ef5c81c25dd99db7236bdee7af371bed5243ac7d122fac33b2bfa794af90fa9f model-00004-of-00021.safetensors
9
+ ba0b18fcc5042786b3f78c766fe62a80166d731dffe135f9c4ce414e6efafe6a model-00005-of-00021.safetensors
10
+ 08bebabb28e0a612a23f6ba5a31cd371c802b1db1e1a28bf3f035625bf084e62 model-00006-of-00021.safetensors
11
+ 9da722882e78a0c8f3c02f38adfef7ab533f221379b35224b37c3ad7eba8703b model-00007-of-00021.safetensors
12
+ c6e8a90c9b9ee8b38a2108486729e90529ed44d37715c882d3e41455b1019931 model-00008-of-00021.safetensors
13
+ 94c2e5dbe9ce10a6ab3b317d8e56d4dd50a28f4ee1e7fa31f3c0d85bb52d6bcb model-00009-of-00021.safetensors
14
+ adf4637af10b0c5183bf26977982559939666eee924771e05bee10696af4f5cc model-00010-of-00021.safetensors
15
+ 5cb73be98675b75c626940ef953a0b7840977c3fe2de667d7cb5570d27ffe334 model-00011-of-00021.safetensors
16
+ 4ca1ca1d8365dfe895e6d9261a697a1120d276a9bc389d748c4b84d69247ec6d model-00012-of-00021.safetensors
17
+ d90c3d131104367e0c7632bdeb2a39e436de76b64e513cad1a0b84302dee53b6 model-00013-of-00021.safetensors
18
+ b757be1e0ca984b34f8044f39e9761f85c254ed799d6379971c4b7a320fbbd6b model-00014-of-00021.safetensors
19
+ 93729dfc86fda8acf9a853d3fb1f3501f7036461d0969ff202d5e8ce6e7218d4 model-00015-of-00021.safetensors
20
+ 5f4b2d423bc6e09d38ebde829a778736011422bca5e2e152a22a7dbb41d6cc3b model-00016-of-00021.safetensors
21
+ d86e35f03ad04b9acfcff590efebd3eddf8fe2d273eb77ada89e2982e1acb2b0 model-00017-of-00021.safetensors
22
+ 1bbe20ed2b8a38ee7a407f7dc39be688aa5a51cc3923636ed912b694e3d56d90 model-00018-of-00021.safetensors
23
+ 8f1ce4190253921566463f349c3c74589de66ef326a2df27da9c41d1be8a160d model-00019-of-00021.safetensors
24
+ 3543886c9a08190d4d3c590c3c4d619c47229f0bd6f4f22910079eb5d9258ade model-00020-of-00021.safetensors
25
+ 4ad717f15bf3f7f9e6536d9f62a9ca2d90145ba83805016a6fe7b63dd586fb8e model-00021-of-00021.safetensors
26
+ 9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 tokenizer.model
model.safetensors.index.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"metadata": {"mergekit_version": "0.0.4.1"}, "weight_map": {"model.layers.8.self_attn.v_proj.weight": "model-00001-of-00021.safetensors", "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00021.safetensors", "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00021.safetensors", "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00021.safetensors", "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00021.safetensors", "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00021.safetensors", "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00021.safetensors", "model.layers.4.mlp.down_proj.weight": "model-00001-of-00021.safetensors", "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00021.safetensors", "model.layers.4.mlp.up_proj.weight": "model-00001-of-00021.safetensors", "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00021.safetensors", "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00021.safetensors", "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00021.safetensors", "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00021.safetensors", "model.layers.4.input_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.3.mlp.down_proj.weight": "model-00001-of-00021.safetensors", "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00021.safetensors", "model.layers.3.mlp.up_proj.weight": "model-00001-of-00021.safetensors", "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00021.safetensors", "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00021.safetensors", "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00021.safetensors", "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00021.safetensors", "model.layers.3.input_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.2.mlp.down_proj.weight": "model-00001-of-00021.safetensors", "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00021.safetensors", "model.layers.2.mlp.up_proj.weight": "model-00001-of-00021.safetensors", "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00021.safetensors", "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00021.safetensors", "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00021.safetensors", "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00021.safetensors", "model.layers.2.input_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.1.mlp.down_proj.weight": "model-00001-of-00021.safetensors", "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00021.safetensors", "model.layers.1.mlp.up_proj.weight": "model-00001-of-00021.safetensors", "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00021.safetensors", "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00021.safetensors", "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00021.safetensors", "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00021.safetensors", "model.layers.1.input_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.0.mlp.down_proj.weight": "model-00001-of-00021.safetensors", "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00021.safetensors", "model.layers.0.mlp.up_proj.weight": "model-00001-of-00021.safetensors", "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00021.safetensors", "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00021.safetensors", "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00021.safetensors", "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00021.safetensors", "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00021.safetensors", "model.layers.0.input_layernorm.weight": "model-00001-of-00021.safetensors", "model.embed_tokens.weight": "model-00001-of-00021.safetensors", "model.layers.10.mlp.down_proj.weight": "model-00002-of-00021.safetensors", "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00021.safetensors", "model.layers.10.mlp.up_proj.weight": "model-00002-of-00021.safetensors", "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00021.safetensors", "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00021.safetensors", "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00021.safetensors", "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00021.safetensors", "model.layers.10.input_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.9.mlp.down_proj.weight": "model-00002-of-00021.safetensors", "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00021.safetensors", "model.layers.9.mlp.up_proj.weight": "model-00002-of-00021.safetensors", "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00021.safetensors", "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00021.safetensors", "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00021.safetensors", "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00021.safetensors", "model.layers.9.input_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.8.mlp.down_proj.weight": "model-00002-of-00021.safetensors", "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00021.safetensors", "model.layers.8.mlp.up_proj.weight": "model-00002-of-00021.safetensors", "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00021.safetensors", "model.layers.8.input_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.7.mlp.down_proj.weight": "model-00002-of-00021.safetensors", "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00021.safetensors", "model.layers.7.mlp.up_proj.weight": "model-00002-of-00021.safetensors", "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00021.safetensors", "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00021.safetensors", "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00021.safetensors", "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00021.safetensors", "model.layers.7.input_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.6.mlp.down_proj.weight": "model-00002-of-00021.safetensors", "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00021.safetensors", "model.layers.6.mlp.up_proj.weight": "model-00002-of-00021.safetensors", "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00021.safetensors", "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00021.safetensors", "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00021.safetensors", "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00021.safetensors", "model.layers.6.input_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.5.mlp.down_proj.weight": "model-00002-of-00021.safetensors", "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00021.safetensors", "model.layers.5.mlp.up_proj.weight": "model-00002-of-00021.safetensors", "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.5.input_layernorm.weight": "model-00002-of-00021.safetensors", "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00021.safetensors", "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00021.safetensors", "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00021.safetensors", "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00021.safetensors", "model.layers.16.mlp.gate_proj.weight": "model-00003-of-00021.safetensors", "model.layers.16.self_attn.o_proj.weight": "model-00003-of-00021.safetensors", "model.layers.16.self_attn.v_proj.weight": "model-00003-of-00021.safetensors", "model.layers.16.self_attn.k_proj.weight": "model-00003-of-00021.safetensors", "model.layers.16.self_attn.q_proj.weight": "model-00003-of-00021.safetensors", "model.layers.15.mlp.down_proj.weight": "model-00003-of-00021.safetensors", "model.layers.15.mlp.gate_proj.weight": "model-00003-of-00021.safetensors", "model.layers.15.mlp.up_proj.weight": "model-00003-of-00021.safetensors", "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.15.self_attn.o_proj.weight": "model-00003-of-00021.safetensors", "model.layers.15.self_attn.v_proj.weight": "model-00003-of-00021.safetensors", "model.layers.15.self_attn.k_proj.weight": "model-00003-of-00021.safetensors", "model.layers.15.self_attn.q_proj.weight": "model-00003-of-00021.safetensors", "model.layers.15.input_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.14.mlp.down_proj.weight": "model-00003-of-00021.safetensors", "model.layers.14.mlp.gate_proj.weight": "model-00003-of-00021.safetensors", "model.layers.14.mlp.up_proj.weight": "model-00003-of-00021.safetensors", "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00021.safetensors", "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00021.safetensors", "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00021.safetensors", "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00021.safetensors", "model.layers.14.input_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.13.mlp.down_proj.weight": "model-00003-of-00021.safetensors", "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00021.safetensors", "model.layers.13.mlp.up_proj.weight": "model-00003-of-00021.safetensors", "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00021.safetensors", "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00021.safetensors", "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00021.safetensors", "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00021.safetensors", "model.layers.13.input_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.12.mlp.down_proj.weight": "model-00003-of-00021.safetensors", "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00021.safetensors", "model.layers.12.mlp.up_proj.weight": "model-00003-of-00021.safetensors", "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00021.safetensors", "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00021.safetensors", "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00021.safetensors", "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00021.safetensors", "model.layers.12.input_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.11.mlp.down_proj.weight": "model-00003-of-00021.safetensors", "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00021.safetensors", "model.layers.11.mlp.up_proj.weight": "model-00003-of-00021.safetensors", "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00021.safetensors", "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00021.safetensors", "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00021.safetensors", "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00021.safetensors", "model.layers.11.input_layernorm.weight": "model-00003-of-00021.safetensors", "model.layers.44.mlp.down_proj.weight": "model-00003-of-00021.safetensors", "model.layers.24.mlp.down_proj.weight": "model-00004-of-00021.safetensors", "model.layers.42.self_attn.v_proj.weight": "model-00004-of-00021.safetensors", "model.layers.22.self_attn.v_proj.weight": "model-00004-of-00021.safetensors", "model.layers.42.self_attn.k_proj.weight": "model-00004-of-00021.safetensors", "model.layers.22.self_attn.k_proj.weight": "model-00004-of-00021.safetensors", "model.layers.42.self_attn.q_proj.weight": "model-00004-of-00021.safetensors", "model.layers.22.self_attn.q_proj.weight": "model-00004-of-00021.safetensors", "model.layers.41.mlp.down_proj.weight": "model-00004-of-00021.safetensors", "model.layers.21.mlp.down_proj.weight": "model-00004-of-00021.safetensors", "model.layers.41.mlp.gate_proj.weight": "model-00004-of-00021.safetensors", "model.layers.21.mlp.gate_proj.weight": "model-00004-of-00021.safetensors", "model.layers.41.mlp.up_proj.weight": "model-00004-of-00021.safetensors", "model.layers.21.mlp.up_proj.weight": "model-00004-of-00021.safetensors", "model.layers.41.post_attention_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.21.post_attention_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.41.self_attn.o_proj.weight": "model-00004-of-00021.safetensors", "model.layers.21.self_attn.o_proj.weight": "model-00004-of-00021.safetensors", "model.layers.41.self_attn.v_proj.weight": "model-00004-of-00021.safetensors", "model.layers.21.self_attn.v_proj.weight": "model-00004-of-00021.safetensors", "model.layers.41.self_attn.k_proj.weight": "model-00004-of-00021.safetensors", "model.layers.21.self_attn.k_proj.weight": "model-00004-of-00021.safetensors", "model.layers.41.self_attn.q_proj.weight": "model-00004-of-00021.safetensors", "model.layers.21.self_attn.q_proj.weight": "model-00004-of-00021.safetensors", "model.layers.41.input_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.21.input_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.40.mlp.down_proj.weight": "model-00004-of-00021.safetensors", "model.layers.20.mlp.down_proj.weight": "model-00004-of-00021.safetensors", "model.layers.40.mlp.gate_proj.weight": "model-00004-of-00021.safetensors", "model.layers.20.mlp.gate_proj.weight": "model-00004-of-00021.safetensors", "model.layers.40.mlp.up_proj.weight": "model-00004-of-00021.safetensors", "model.layers.20.mlp.up_proj.weight": "model-00004-of-00021.safetensors", "model.layers.40.post_attention_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.40.self_attn.o_proj.weight": "model-00004-of-00021.safetensors", "model.layers.20.self_attn.o_proj.weight": "model-00004-of-00021.safetensors", "model.layers.40.self_attn.v_proj.weight": "model-00004-of-00021.safetensors", "model.layers.20.self_attn.v_proj.weight": "model-00004-of-00021.safetensors", "model.layers.40.self_attn.k_proj.weight": "model-00004-of-00021.safetensors", "model.layers.20.self_attn.k_proj.weight": "model-00004-of-00021.safetensors", "model.layers.40.self_attn.q_proj.weight": "model-00004-of-00021.safetensors", "model.layers.20.self_attn.q_proj.weight": "model-00004-of-00021.safetensors", "model.layers.40.input_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.20.input_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.19.mlp.down_proj.weight": "model-00004-of-00021.safetensors", "model.layers.19.mlp.gate_proj.weight": "model-00004-of-00021.safetensors", "model.layers.19.mlp.up_proj.weight": "model-00004-of-00021.safetensors", "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.19.input_layernorm.weight": "model-00004-of-00021.safetensors", "model.layers.18.mlp.down_proj.weight": "model-00004-of-00021.safetensors", "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00021.safetensors", "model.layers.18.mlp.up_proj.weight": "model-00005-of-00021.safetensors", "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00021.safetensors", "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00021.safetensors", "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00021.safetensors", "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00021.safetensors", "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00021.safetensors", "model.layers.18.input_layernorm.weight": "model-00005-of-00021.safetensors", "model.layers.17.mlp.down_proj.weight": "model-00005-of-00021.safetensors", "model.layers.17.mlp.gate_proj.weight": "model-00005-of-00021.safetensors", "model.layers.17.mlp.up_proj.weight": "model-00005-of-00021.safetensors", "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00021.safetensors", "model.layers.17.self_attn.o_proj.weight": "model-00005-of-00021.safetensors", "model.layers.17.self_attn.v_proj.weight": "model-00005-of-00021.safetensors", "model.layers.17.self_attn.k_proj.weight": "model-00005-of-00021.safetensors", "model.layers.17.self_attn.q_proj.weight": "model-00005-of-00021.safetensors", "model.layers.17.input_layernorm.weight": "model-00005-of-00021.safetensors", "model.layers.16.mlp.down_proj.weight": "model-00005-of-00021.safetensors", "model.layers.16.mlp.up_proj.weight": "model-00005-of-00021.safetensors", "model.layers.16.post_attention_layernorm.weight": "model-00005-of-00021.safetensors", "model.layers.16.input_layernorm.weight": "model-00005-of-00021.safetensors", "model.layers.50.mlp.gate_proj.weight": "model-00005-of-00021.safetensors", "model.layers.30.mlp.gate_proj.weight": "model-00005-of-00021.safetensors", "model.layers.50.self_attn.o_proj.weight": "model-00005-of-00021.safetensors", "model.layers.30.self_attn.o_proj.weight": "model-00005-of-00021.safetensors", "model.layers.50.self_attn.v_proj.weight": "model-00005-of-00021.safetensors", "model.layers.30.self_attn.v_proj.weight": "model-00005-of-00021.safetensors", "model.layers.50.self_attn.k_proj.weight": "model-00005-of-00021.safetensors", "model.layers.30.self_attn.k_proj.weight": "model-00005-of-00021.safetensors", "model.layers.50.self_attn.q_proj.weight": "model-00005-of-00021.safetensors", "model.layers.30.self_attn.q_proj.weight": "model-00005-of-00021.safetensors", "model.layers.47.mlp.gate_proj.weight": "model-00005-of-00021.safetensors", "model.layers.27.mlp.gate_proj.weight": "model-00005-of-00021.safetensors", "model.layers.47.mlp.up_proj.weight": "model-00005-of-00021.safetensors", "model.layers.27.mlp.up_proj.weight": "model-00005-of-00021.safetensors", "model.layers.47.self_attn.o_proj.weight": "model-00005-of-00021.safetensors", "model.layers.27.self_attn.o_proj.weight": "model-00005-of-00021.safetensors", "model.layers.47.self_attn.v_proj.weight": "model-00005-of-00021.safetensors", "model.layers.27.self_attn.v_proj.weight": "model-00005-of-00021.safetensors", "model.layers.47.self_attn.k_proj.weight": "model-00005-of-00021.safetensors", "model.layers.27.self_attn.k_proj.weight": "model-00005-of-00021.safetensors", "model.layers.47.self_attn.q_proj.weight": "model-00005-of-00021.safetensors", "model.layers.27.self_attn.q_proj.weight": "model-00005-of-00021.safetensors", "model.layers.46.mlp.down_proj.weight": "model-00005-of-00021.safetensors", "model.layers.26.mlp.down_proj.weight": "model-00005-of-00021.safetensors", "model.layers.46.mlp.gate_proj.weight": "model-00005-of-00021.safetensors", "model.layers.26.mlp.gate_proj.weight": "model-00005-of-00021.safetensors", "model.layers.46.mlp.up_proj.weight": "model-00005-of-00021.safetensors", "model.layers.26.mlp.up_proj.weight": "model-00006-of-00021.safetensors", "model.layers.46.post_attention_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.46.self_attn.o_proj.weight": "model-00006-of-00021.safetensors", "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00021.safetensors", "model.layers.46.self_attn.v_proj.weight": "model-00006-of-00021.safetensors", "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00021.safetensors", "model.layers.46.self_attn.k_proj.weight": "model-00006-of-00021.safetensors", "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00021.safetensors", "model.layers.46.self_attn.q_proj.weight": "model-00006-of-00021.safetensors", "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00021.safetensors", "model.layers.46.input_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.26.input_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.45.mlp.down_proj.weight": "model-00006-of-00021.safetensors", "model.layers.25.mlp.down_proj.weight": "model-00006-of-00021.safetensors", "model.layers.45.mlp.gate_proj.weight": "model-00006-of-00021.safetensors", "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00021.safetensors", "model.layers.45.mlp.up_proj.weight": "model-00006-of-00021.safetensors", "model.layers.25.mlp.up_proj.weight": "model-00006-of-00021.safetensors", "model.layers.45.post_attention_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.45.self_attn.o_proj.weight": "model-00006-of-00021.safetensors", "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00021.safetensors", "model.layers.45.self_attn.v_proj.weight": "model-00006-of-00021.safetensors", "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00021.safetensors", "model.layers.45.self_attn.k_proj.weight": "model-00006-of-00021.safetensors", "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00021.safetensors", "model.layers.45.self_attn.q_proj.weight": "model-00006-of-00021.safetensors", "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00021.safetensors", "model.layers.45.input_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.25.input_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.44.mlp.gate_proj.weight": "model-00006-of-00021.safetensors", "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00021.safetensors", "model.layers.44.mlp.up_proj.weight": "model-00006-of-00021.safetensors", "model.layers.24.mlp.up_proj.weight": "model-00006-of-00021.safetensors", "model.layers.44.post_attention_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.44.self_attn.o_proj.weight": "model-00006-of-00021.safetensors", "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00021.safetensors", "model.layers.44.self_attn.v_proj.weight": "model-00006-of-00021.safetensors", "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00021.safetensors", "model.layers.44.self_attn.k_proj.weight": "model-00006-of-00021.safetensors", "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00021.safetensors", "model.layers.44.self_attn.q_proj.weight": "model-00006-of-00021.safetensors", "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00021.safetensors", "model.layers.44.input_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.24.input_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.43.mlp.down_proj.weight": "model-00006-of-00021.safetensors", "model.layers.23.mlp.down_proj.weight": "model-00006-of-00021.safetensors", "model.layers.43.mlp.gate_proj.weight": "model-00006-of-00021.safetensors", "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00021.safetensors", "model.layers.43.mlp.up_proj.weight": "model-00006-of-00021.safetensors", "model.layers.23.mlp.up_proj.weight": "model-00006-of-00021.safetensors", "model.layers.43.post_attention_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00021.safetensors", "model.layers.43.self_attn.o_proj.weight": "model-00006-of-00021.safetensors", "model.layers.23.self_attn.o_proj.weight": "model-00007-of-00021.safetensors", "model.layers.43.self_attn.v_proj.weight": "model-00007-of-00021.safetensors", "model.layers.23.self_attn.v_proj.weight": "model-00007-of-00021.safetensors", "model.layers.43.self_attn.k_proj.weight": "model-00007-of-00021.safetensors", "model.layers.23.self_attn.k_proj.weight": "model-00007-of-00021.safetensors", "model.layers.43.self_attn.q_proj.weight": "model-00007-of-00021.safetensors", "model.layers.23.self_attn.q_proj.weight": "model-00007-of-00021.safetensors", "model.layers.43.input_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.23.input_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.42.mlp.down_proj.weight": "model-00007-of-00021.safetensors", "model.layers.22.mlp.down_proj.weight": "model-00007-of-00021.safetensors", "model.layers.42.mlp.gate_proj.weight": "model-00007-of-00021.safetensors", "model.layers.22.mlp.gate_proj.weight": "model-00007-of-00021.safetensors", "model.layers.42.mlp.up_proj.weight": "model-00007-of-00021.safetensors", "model.layers.22.mlp.up_proj.weight": "model-00007-of-00021.safetensors", "model.layers.42.post_attention_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.22.post_attention_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.42.self_attn.o_proj.weight": "model-00007-of-00021.safetensors", "model.layers.22.self_attn.o_proj.weight": "model-00007-of-00021.safetensors", "model.layers.42.input_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.22.input_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.56.self_attn.v_proj.weight": "model-00007-of-00021.safetensors", "model.layers.36.self_attn.v_proj.weight": "model-00007-of-00021.safetensors", "model.layers.56.self_attn.k_proj.weight": "model-00007-of-00021.safetensors", "model.layers.36.self_attn.k_proj.weight": "model-00007-of-00021.safetensors", "model.layers.56.self_attn.q_proj.weight": "model-00007-of-00021.safetensors", "model.layers.36.self_attn.q_proj.weight": "model-00007-of-00021.safetensors", "model.layers.55.mlp.down_proj.weight": "model-00007-of-00021.safetensors", "model.layers.35.mlp.down_proj.weight": "model-00007-of-00021.safetensors", "model.layers.55.mlp.gate_proj.weight": "model-00007-of-00021.safetensors", "model.layers.35.mlp.gate_proj.weight": "model-00007-of-00021.safetensors", "model.layers.53.self_attn.o_proj.weight": "model-00007-of-00021.safetensors", "model.layers.33.self_attn.o_proj.weight": "model-00007-of-00021.safetensors", "model.layers.53.self_attn.v_proj.weight": "model-00007-of-00021.safetensors", "model.layers.33.self_attn.v_proj.weight": "model-00007-of-00021.safetensors", "model.layers.53.self_attn.k_proj.weight": "model-00007-of-00021.safetensors", "model.layers.33.self_attn.k_proj.weight": "model-00007-of-00021.safetensors", "model.layers.53.self_attn.q_proj.weight": "model-00007-of-00021.safetensors", "model.layers.33.self_attn.q_proj.weight": "model-00007-of-00021.safetensors", "model.layers.52.mlp.down_proj.weight": "model-00007-of-00021.safetensors", "model.layers.32.mlp.down_proj.weight": "model-00007-of-00021.safetensors", "model.layers.52.mlp.gate_proj.weight": "model-00007-of-00021.safetensors", "model.layers.32.mlp.gate_proj.weight": "model-00007-of-00021.safetensors", "model.layers.52.mlp.up_proj.weight": "model-00007-of-00021.safetensors", "model.layers.32.mlp.up_proj.weight": "model-00007-of-00021.safetensors", "model.layers.52.post_attention_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.32.post_attention_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.52.self_attn.o_proj.weight": "model-00007-of-00021.safetensors", "model.layers.32.self_attn.o_proj.weight": "model-00007-of-00021.safetensors", "model.layers.52.self_attn.v_proj.weight": "model-00007-of-00021.safetensors", "model.layers.32.self_attn.v_proj.weight": "model-00007-of-00021.safetensors", "model.layers.52.self_attn.k_proj.weight": "model-00007-of-00021.safetensors", "model.layers.32.self_attn.k_proj.weight": "model-00007-of-00021.safetensors", "model.layers.52.self_attn.q_proj.weight": "model-00007-of-00021.safetensors", "model.layers.32.self_attn.q_proj.weight": "model-00007-of-00021.safetensors", "model.layers.52.input_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.32.input_layernorm.weight": "model-00007-of-00021.safetensors", "model.layers.51.mlp.down_proj.weight": "model-00008-of-00021.safetensors", "model.layers.31.mlp.down_proj.weight": "model-00008-of-00021.safetensors", "model.layers.51.mlp.gate_proj.weight": "model-00008-of-00021.safetensors", "model.layers.31.mlp.gate_proj.weight": "model-00008-of-00021.safetensors", "model.layers.51.mlp.up_proj.weight": "model-00008-of-00021.safetensors", "model.layers.31.mlp.up_proj.weight": "model-00008-of-00021.safetensors", "model.layers.51.post_attention_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.31.post_attention_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.51.self_attn.o_proj.weight": "model-00008-of-00021.safetensors", "model.layers.31.self_attn.o_proj.weight": "model-00008-of-00021.safetensors", "model.layers.51.self_attn.v_proj.weight": "model-00008-of-00021.safetensors", "model.layers.31.self_attn.v_proj.weight": "model-00008-of-00021.safetensors", "model.layers.51.self_attn.k_proj.weight": "model-00008-of-00021.safetensors", "model.layers.31.self_attn.k_proj.weight": "model-00008-of-00021.safetensors", "model.layers.51.self_attn.q_proj.weight": "model-00008-of-00021.safetensors", "model.layers.31.self_attn.q_proj.weight": "model-00008-of-00021.safetensors", "model.layers.51.input_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.31.input_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.50.mlp.down_proj.weight": "model-00008-of-00021.safetensors", "model.layers.30.mlp.down_proj.weight": "model-00008-of-00021.safetensors", "model.layers.50.mlp.up_proj.weight": "model-00008-of-00021.safetensors", "model.layers.30.mlp.up_proj.weight": "model-00008-of-00021.safetensors", "model.layers.50.post_attention_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.30.post_attention_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.50.input_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.30.input_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.49.mlp.down_proj.weight": "model-00008-of-00021.safetensors", "model.layers.29.mlp.down_proj.weight": "model-00008-of-00021.safetensors", "model.layers.49.mlp.gate_proj.weight": "model-00008-of-00021.safetensors", "model.layers.29.mlp.gate_proj.weight": "model-00008-of-00021.safetensors", "model.layers.49.mlp.up_proj.weight": "model-00008-of-00021.safetensors", "model.layers.29.mlp.up_proj.weight": "model-00008-of-00021.safetensors", "model.layers.49.post_attention_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.29.post_attention_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.49.self_attn.o_proj.weight": "model-00008-of-00021.safetensors", "model.layers.29.self_attn.o_proj.weight": "model-00008-of-00021.safetensors", "model.layers.49.self_attn.v_proj.weight": "model-00008-of-00021.safetensors", "model.layers.29.self_attn.v_proj.weight": "model-00008-of-00021.safetensors", "model.layers.49.self_attn.k_proj.weight": "model-00008-of-00021.safetensors", "model.layers.29.self_attn.k_proj.weight": "model-00008-of-00021.safetensors", "model.layers.49.self_attn.q_proj.weight": "model-00008-of-00021.safetensors", "model.layers.29.self_attn.q_proj.weight": "model-00008-of-00021.safetensors", "model.layers.49.input_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.29.input_layernorm.weight": "model-00008-of-00021.safetensors", "model.layers.48.mlp.down_proj.weight": "model-00008-of-00021.safetensors", "model.layers.28.mlp.down_proj.weight": "model-00008-of-00021.safetensors", "model.layers.48.mlp.gate_proj.weight": "model-00009-of-00021.safetensors", "model.layers.28.mlp.gate_proj.weight": "model-00009-of-00021.safetensors", "model.layers.48.mlp.up_proj.weight": "model-00009-of-00021.safetensors", "model.layers.28.mlp.up_proj.weight": "model-00009-of-00021.safetensors", "model.layers.48.post_attention_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.28.post_attention_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.48.self_attn.o_proj.weight": "model-00009-of-00021.safetensors", "model.layers.28.self_attn.o_proj.weight": "model-00009-of-00021.safetensors", "model.layers.48.self_attn.v_proj.weight": "model-00009-of-00021.safetensors", "model.layers.28.self_attn.v_proj.weight": "model-00009-of-00021.safetensors", "model.layers.48.self_attn.k_proj.weight": "model-00009-of-00021.safetensors", "model.layers.28.self_attn.k_proj.weight": "model-00009-of-00021.safetensors", "model.layers.48.self_attn.q_proj.weight": "model-00009-of-00021.safetensors", "model.layers.28.self_attn.q_proj.weight": "model-00009-of-00021.safetensors", "model.layers.48.input_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.28.input_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.47.mlp.down_proj.weight": "model-00009-of-00021.safetensors", "model.layers.27.mlp.down_proj.weight": "model-00009-of-00021.safetensors", "model.layers.47.post_attention_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.27.post_attention_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.47.input_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.27.input_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.81.mlp.gate_proj.weight": "model-00009-of-00021.safetensors", "model.layers.61.mlp.gate_proj.weight": "model-00009-of-00021.safetensors", "model.layers.81.mlp.up_proj.weight": "model-00009-of-00021.safetensors", "model.layers.61.mlp.up_proj.weight": "model-00009-of-00021.safetensors", "model.layers.81.self_attn.o_proj.weight": "model-00009-of-00021.safetensors", "model.layers.61.self_attn.o_proj.weight": "model-00009-of-00021.safetensors", "model.layers.81.self_attn.v_proj.weight": "model-00009-of-00021.safetensors", "model.layers.61.self_attn.v_proj.weight": "model-00009-of-00021.safetensors", "model.layers.81.self_attn.k_proj.weight": "model-00009-of-00021.safetensors", "model.layers.61.self_attn.k_proj.weight": "model-00009-of-00021.safetensors", "model.layers.58.mlp.down_proj.weight": "model-00009-of-00021.safetensors", "model.layers.38.mlp.down_proj.weight": "model-00009-of-00021.safetensors", "model.layers.58.mlp.gate_proj.weight": "model-00009-of-00021.safetensors", "model.layers.38.mlp.gate_proj.weight": "model-00009-of-00021.safetensors", "model.layers.58.mlp.up_proj.weight": "model-00009-of-00021.safetensors", "model.layers.38.mlp.up_proj.weight": "model-00009-of-00021.safetensors", "model.layers.58.post_attention_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.38.post_attention_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.58.self_attn.o_proj.weight": "model-00009-of-00021.safetensors", "model.layers.38.self_attn.o_proj.weight": "model-00009-of-00021.safetensors", "model.layers.58.self_attn.v_proj.weight": "model-00009-of-00021.safetensors", "model.layers.38.self_attn.v_proj.weight": "model-00009-of-00021.safetensors", "model.layers.58.self_attn.k_proj.weight": "model-00009-of-00021.safetensors", "model.layers.38.self_attn.k_proj.weight": "model-00009-of-00021.safetensors", "model.layers.58.self_attn.q_proj.weight": "model-00009-of-00021.safetensors", "model.layers.38.self_attn.q_proj.weight": "model-00009-of-00021.safetensors", "model.layers.58.input_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.38.input_layernorm.weight": "model-00009-of-00021.safetensors", "model.layers.57.mlp.down_proj.weight": "model-00009-of-00021.safetensors", "model.layers.37.mlp.down_proj.weight": "model-00009-of-00021.safetensors", "model.layers.57.mlp.gate_proj.weight": "model-00010-of-00021.safetensors", "model.layers.37.mlp.gate_proj.weight": "model-00010-of-00021.safetensors", "model.layers.57.mlp.up_proj.weight": "model-00010-of-00021.safetensors", "model.layers.37.mlp.up_proj.weight": "model-00010-of-00021.safetensors", "model.layers.57.post_attention_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.37.post_attention_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.57.self_attn.o_proj.weight": "model-00010-of-00021.safetensors", "model.layers.37.self_attn.o_proj.weight": "model-00010-of-00021.safetensors", "model.layers.57.self_attn.v_proj.weight": "model-00010-of-00021.safetensors", "model.layers.37.self_attn.v_proj.weight": "model-00010-of-00021.safetensors", "model.layers.57.self_attn.k_proj.weight": "model-00010-of-00021.safetensors", "model.layers.37.self_attn.k_proj.weight": "model-00010-of-00021.safetensors", "model.layers.57.self_attn.q_proj.weight": "model-00010-of-00021.safetensors", "model.layers.37.self_attn.q_proj.weight": "model-00010-of-00021.safetensors", "model.layers.57.input_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.37.input_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.56.mlp.down_proj.weight": "model-00010-of-00021.safetensors", "model.layers.36.mlp.down_proj.weight": "model-00010-of-00021.safetensors", "model.layers.56.mlp.gate_proj.weight": "model-00010-of-00021.safetensors", "model.layers.36.mlp.gate_proj.weight": "model-00010-of-00021.safetensors", "model.layers.56.mlp.up_proj.weight": "model-00010-of-00021.safetensors", "model.layers.36.mlp.up_proj.weight": "model-00010-of-00021.safetensors", "model.layers.56.post_attention_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.36.post_attention_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.56.self_attn.o_proj.weight": "model-00010-of-00021.safetensors", "model.layers.36.self_attn.o_proj.weight": "model-00010-of-00021.safetensors", "model.layers.56.input_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.36.input_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.55.mlp.up_proj.weight": "model-00010-of-00021.safetensors", "model.layers.35.mlp.up_proj.weight": "model-00010-of-00021.safetensors", "model.layers.55.post_attention_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.35.post_attention_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.55.self_attn.o_proj.weight": "model-00010-of-00021.safetensors", "model.layers.35.self_attn.o_proj.weight": "model-00010-of-00021.safetensors", "model.layers.55.self_attn.v_proj.weight": "model-00010-of-00021.safetensors", "model.layers.35.self_attn.v_proj.weight": "model-00010-of-00021.safetensors", "model.layers.55.self_attn.k_proj.weight": "model-00010-of-00021.safetensors", "model.layers.35.self_attn.k_proj.weight": "model-00010-of-00021.safetensors", "model.layers.55.self_attn.q_proj.weight": "model-00010-of-00021.safetensors", "model.layers.35.self_attn.q_proj.weight": "model-00010-of-00021.safetensors", "model.layers.55.input_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.35.input_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.54.mlp.down_proj.weight": "model-00010-of-00021.safetensors", "model.layers.34.mlp.down_proj.weight": "model-00010-of-00021.safetensors", "model.layers.54.mlp.gate_proj.weight": "model-00010-of-00021.safetensors", "model.layers.34.mlp.gate_proj.weight": "model-00010-of-00021.safetensors", "model.layers.54.mlp.up_proj.weight": "model-00010-of-00021.safetensors", "model.layers.34.mlp.up_proj.weight": "model-00010-of-00021.safetensors", "model.layers.54.post_attention_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.34.post_attention_layernorm.weight": "model-00010-of-00021.safetensors", "model.layers.54.self_attn.o_proj.weight": "model-00011-of-00021.safetensors", "model.layers.34.self_attn.o_proj.weight": "model-00011-of-00021.safetensors", "model.layers.54.self_attn.v_proj.weight": "model-00011-of-00021.safetensors", "model.layers.34.self_attn.v_proj.weight": "model-00011-of-00021.safetensors", "model.layers.54.self_attn.k_proj.weight": "model-00011-of-00021.safetensors", "model.layers.34.self_attn.k_proj.weight": "model-00011-of-00021.safetensors", "model.layers.54.self_attn.q_proj.weight": "model-00011-of-00021.safetensors", "model.layers.34.self_attn.q_proj.weight": "model-00011-of-00021.safetensors", "model.layers.54.input_layernorm.weight": "model-00011-of-00021.safetensors", "model.layers.34.input_layernorm.weight": "model-00011-of-00021.safetensors", "model.layers.53.mlp.down_proj.weight": "model-00011-of-00021.safetensors", "model.layers.33.mlp.down_proj.weight": "model-00011-of-00021.safetensors", "model.layers.53.mlp.gate_proj.weight": "model-00011-of-00021.safetensors", "model.layers.33.mlp.gate_proj.weight": "model-00011-of-00021.safetensors", "model.layers.53.mlp.up_proj.weight": "model-00011-of-00021.safetensors", "model.layers.33.mlp.up_proj.weight": "model-00011-of-00021.safetensors", "model.layers.53.post_attention_layernorm.weight": "model-00011-of-00021.safetensors", "model.layers.33.post_attention_layernorm.weight": "model-00011-of-00021.safetensors", "model.layers.53.input_layernorm.weight": "model-00011-of-00021.safetensors", "model.layers.33.input_layernorm.weight": "model-00011-of-00021.safetensors", "model.layers.87.self_attn.o_proj.weight": "model-00011-of-00021.safetensors", "model.layers.67.self_attn.o_proj.weight": "model-00011-of-00021.safetensors", "model.layers.87.self_attn.v_proj.weight": "model-00011-of-00021.safetensors", "model.layers.67.self_attn.v_proj.weight": "model-00011-of-00021.safetensors", "model.layers.87.self_attn.k_proj.weight": "model-00011-of-00021.safetensors", "model.layers.67.self_attn.k_proj.weight": "model-00011-of-00021.safetensors", "model.layers.87.self_attn.q_proj.weight": "model-00011-of-00021.safetensors", "model.layers.67.self_attn.q_proj.weight": "model-00011-of-00021.safetensors", "model.layers.86.mlp.down_proj.weight": "model-00011-of-00021.safetensors", "model.layers.66.mlp.down_proj.weight": "model-00011-of-00021.safetensors", "model.layers.86.mlp.gate_proj.weight": "model-00011-of-00021.safetensors", "model.layers.66.mlp.gate_proj.weight": "model-00011-of-00021.safetensors", "model.layers.86.mlp.up_proj.weight": "model-00011-of-00021.safetensors", "model.layers.66.mlp.up_proj.weight": "model-00011-of-00021.safetensors", "model.layers.86.post_attention_layernorm.weight": "model-00011-of-00021.safetensors", "model.layers.66.post_attention_layernorm.weight": "model-00011-of-00021.safetensors", "model.layers.84.mlp.gate_proj.weight": "model-00011-of-00021.safetensors", "model.layers.64.mlp.gate_proj.weight": "model-00011-of-00021.safetensors", "model.layers.84.self_attn.o_proj.weight": "model-00011-of-00021.safetensors", "model.layers.64.self_attn.o_proj.weight": "model-00011-of-00021.safetensors", "model.layers.84.self_attn.v_proj.weight": "model-00011-of-00021.safetensors", "model.layers.64.self_attn.v_proj.weight": "model-00011-of-00021.safetensors", "model.layers.84.self_attn.k_proj.weight": "model-00011-of-00021.safetensors", "model.layers.64.self_attn.k_proj.weight": "model-00011-of-00021.safetensors", "model.layers.84.self_attn.q_proj.weight": "model-00011-of-00021.safetensors", "model.layers.64.self_attn.q_proj.weight": "model-00011-of-00021.safetensors", "model.layers.83.mlp.down_proj.weight": "model-00011-of-00021.safetensors", "model.layers.63.mlp.down_proj.weight": "model-00011-of-00021.safetensors", "model.layers.83.mlp.gate_proj.weight": "model-00011-of-00021.safetensors", "model.layers.63.mlp.gate_proj.weight": "model-00012-of-00021.safetensors", "model.layers.83.mlp.up_proj.weight": "model-00012-of-00021.safetensors", "model.layers.63.mlp.up_proj.weight": "model-00012-of-00021.safetensors", "model.layers.83.post_attention_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.63.post_attention_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.83.self_attn.o_proj.weight": "model-00012-of-00021.safetensors", "model.layers.63.self_attn.o_proj.weight": "model-00012-of-00021.safetensors", "model.layers.83.self_attn.v_proj.weight": "model-00012-of-00021.safetensors", "model.layers.63.self_attn.v_proj.weight": "model-00012-of-00021.safetensors", "model.layers.83.self_attn.k_proj.weight": "model-00012-of-00021.safetensors", "model.layers.63.self_attn.k_proj.weight": "model-00012-of-00021.safetensors", "model.layers.83.self_attn.q_proj.weight": "model-00012-of-00021.safetensors", "model.layers.63.self_attn.q_proj.weight": "model-00012-of-00021.safetensors", "model.layers.83.input_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.63.input_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.82.mlp.down_proj.weight": "model-00012-of-00021.safetensors", "model.layers.62.mlp.down_proj.weight": "model-00012-of-00021.safetensors", "model.layers.82.mlp.gate_proj.weight": "model-00012-of-00021.safetensors", "model.layers.62.mlp.gate_proj.weight": "model-00012-of-00021.safetensors", "model.layers.82.mlp.up_proj.weight": "model-00012-of-00021.safetensors", "model.layers.62.mlp.up_proj.weight": "model-00012-of-00021.safetensors", "model.layers.82.post_attention_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.62.post_attention_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.82.self_attn.o_proj.weight": "model-00012-of-00021.safetensors", "model.layers.62.self_attn.o_proj.weight": "model-00012-of-00021.safetensors", "model.layers.82.self_attn.v_proj.weight": "model-00012-of-00021.safetensors", "model.layers.62.self_attn.v_proj.weight": "model-00012-of-00021.safetensors", "model.layers.82.self_attn.k_proj.weight": "model-00012-of-00021.safetensors", "model.layers.62.self_attn.k_proj.weight": "model-00012-of-00021.safetensors", "model.layers.82.self_attn.q_proj.weight": "model-00012-of-00021.safetensors", "model.layers.62.self_attn.q_proj.weight": "model-00012-of-00021.safetensors", "model.layers.82.input_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.62.input_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.81.mlp.down_proj.weight": "model-00012-of-00021.safetensors", "model.layers.61.mlp.down_proj.weight": "model-00012-of-00021.safetensors", "model.layers.81.post_attention_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.61.post_attention_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.81.self_attn.q_proj.weight": "model-00012-of-00021.safetensors", "model.layers.61.self_attn.q_proj.weight": "model-00012-of-00021.safetensors", "model.layers.81.input_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.61.input_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.80.mlp.down_proj.weight": "model-00012-of-00021.safetensors", "model.layers.60.mlp.down_proj.weight": "model-00012-of-00021.safetensors", "model.layers.80.mlp.gate_proj.weight": "model-00012-of-00021.safetensors", "model.layers.60.mlp.gate_proj.weight": "model-00012-of-00021.safetensors", "model.layers.80.mlp.up_proj.weight": "model-00012-of-00021.safetensors", "model.layers.60.mlp.up_proj.weight": "model-00012-of-00021.safetensors", "model.layers.80.post_attention_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.60.post_attention_layernorm.weight": "model-00012-of-00021.safetensors", "model.layers.80.self_attn.o_proj.weight": "model-00012-of-00021.safetensors", "model.layers.60.self_attn.o_proj.weight": "model-00012-of-00021.safetensors", "model.layers.80.self_attn.v_proj.weight": "model-00012-of-00021.safetensors", "model.layers.60.self_attn.v_proj.weight": "model-00012-of-00021.safetensors", "model.layers.80.self_attn.k_proj.weight": "model-00012-of-00021.safetensors", "model.layers.60.self_attn.k_proj.weight": "model-00012-of-00021.safetensors", "model.layers.80.self_attn.q_proj.weight": "model-00012-of-00021.safetensors", "model.layers.60.self_attn.q_proj.weight": "model-00013-of-00021.safetensors", "model.layers.80.input_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.60.input_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.59.mlp.down_proj.weight": "model-00013-of-00021.safetensors", "model.layers.39.mlp.down_proj.weight": "model-00013-of-00021.safetensors", "model.layers.59.mlp.gate_proj.weight": "model-00013-of-00021.safetensors", "model.layers.39.mlp.gate_proj.weight": "model-00013-of-00021.safetensors", "model.layers.59.mlp.up_proj.weight": "model-00013-of-00021.safetensors", "model.layers.39.mlp.up_proj.weight": "model-00013-of-00021.safetensors", "model.layers.59.post_attention_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.39.post_attention_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.59.self_attn.o_proj.weight": "model-00013-of-00021.safetensors", "model.layers.39.self_attn.o_proj.weight": "model-00013-of-00021.safetensors", "model.layers.59.self_attn.v_proj.weight": "model-00013-of-00021.safetensors", "model.layers.39.self_attn.v_proj.weight": "model-00013-of-00021.safetensors", "model.layers.59.self_attn.k_proj.weight": "model-00013-of-00021.safetensors", "model.layers.39.self_attn.k_proj.weight": "model-00013-of-00021.safetensors", "model.layers.59.self_attn.q_proj.weight": "model-00013-of-00021.safetensors", "model.layers.39.self_attn.q_proj.weight": "model-00013-of-00021.safetensors", "model.layers.59.input_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.39.input_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.92.mlp.down_proj.weight": "model-00013-of-00021.safetensors", "model.layers.72.mlp.down_proj.weight": "model-00013-of-00021.safetensors", "model.layers.92.mlp.gate_proj.weight": "model-00013-of-00021.safetensors", "model.layers.72.mlp.gate_proj.weight": "model-00013-of-00021.safetensors", "model.layers.92.mlp.up_proj.weight": "model-00013-of-00021.safetensors", "model.layers.72.mlp.up_proj.weight": "model-00013-of-00021.safetensors", "model.layers.92.post_attention_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.72.post_attention_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.92.self_attn.o_proj.weight": "model-00013-of-00021.safetensors", "model.layers.72.self_attn.o_proj.weight": "model-00013-of-00021.safetensors", "model.layers.92.self_attn.v_proj.weight": "model-00013-of-00021.safetensors", "model.layers.72.self_attn.v_proj.weight": "model-00013-of-00021.safetensors", "model.layers.92.self_attn.k_proj.weight": "model-00013-of-00021.safetensors", "model.layers.72.self_attn.k_proj.weight": "model-00013-of-00021.safetensors", "model.layers.92.self_attn.q_proj.weight": "model-00013-of-00021.safetensors", "model.layers.72.self_attn.q_proj.weight": "model-00013-of-00021.safetensors", "model.layers.92.input_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.72.input_layernorm.weight": "model-00013-of-00021.safetensors", "model.layers.90.self_attn.v_proj.weight": "model-00013-of-00021.safetensors", "model.layers.70.self_attn.v_proj.weight": "model-00013-of-00021.safetensors", "model.layers.90.self_attn.k_proj.weight": "model-00013-of-00021.safetensors", "model.layers.70.self_attn.k_proj.weight": "model-00013-of-00021.safetensors", "model.layers.90.self_attn.q_proj.weight": "model-00013-of-00021.safetensors", "model.layers.70.self_attn.q_proj.weight": "model-00013-of-00021.safetensors", "model.layers.89.mlp.down_proj.weight": "model-00013-of-00021.safetensors", "model.layers.69.mlp.down_proj.weight": "model-00013-of-00021.safetensors", "model.layers.89.mlp.gate_proj.weight": "model-00013-of-00021.safetensors", "model.layers.69.mlp.gate_proj.weight": "model-00013-of-00021.safetensors", "model.layers.89.mlp.up_proj.weight": "model-00013-of-00021.safetensors", "model.layers.69.mlp.up_proj.weight": "model-00014-of-00021.safetensors", "model.layers.89.post_attention_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.69.post_attention_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.89.self_attn.o_proj.weight": "model-00014-of-00021.safetensors", "model.layers.69.self_attn.o_proj.weight": "model-00014-of-00021.safetensors", "model.layers.89.self_attn.v_proj.weight": "model-00014-of-00021.safetensors", "model.layers.69.self_attn.v_proj.weight": "model-00014-of-00021.safetensors", "model.layers.89.self_attn.k_proj.weight": "model-00014-of-00021.safetensors", "model.layers.69.self_attn.k_proj.weight": "model-00014-of-00021.safetensors", "model.layers.89.self_attn.q_proj.weight": "model-00014-of-00021.safetensors", "model.layers.69.self_attn.q_proj.weight": "model-00014-of-00021.safetensors", "model.layers.89.input_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.69.input_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.88.mlp.down_proj.weight": "model-00014-of-00021.safetensors", "model.layers.68.mlp.down_proj.weight": "model-00014-of-00021.safetensors", "model.layers.88.mlp.gate_proj.weight": "model-00014-of-00021.safetensors", "model.layers.68.mlp.gate_proj.weight": "model-00014-of-00021.safetensors", "model.layers.88.mlp.up_proj.weight": "model-00014-of-00021.safetensors", "model.layers.68.mlp.up_proj.weight": "model-00014-of-00021.safetensors", "model.layers.88.post_attention_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.68.post_attention_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.88.self_attn.o_proj.weight": "model-00014-of-00021.safetensors", "model.layers.68.self_attn.o_proj.weight": "model-00014-of-00021.safetensors", "model.layers.88.self_attn.v_proj.weight": "model-00014-of-00021.safetensors", "model.layers.68.self_attn.v_proj.weight": "model-00014-of-00021.safetensors", "model.layers.88.self_attn.k_proj.weight": "model-00014-of-00021.safetensors", "model.layers.68.self_attn.k_proj.weight": "model-00014-of-00021.safetensors", "model.layers.88.self_attn.q_proj.weight": "model-00014-of-00021.safetensors", "model.layers.68.self_attn.q_proj.weight": "model-00014-of-00021.safetensors", "model.layers.88.input_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.68.input_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.87.mlp.down_proj.weight": "model-00014-of-00021.safetensors", "model.layers.67.mlp.down_proj.weight": "model-00014-of-00021.safetensors", "model.layers.87.mlp.gate_proj.weight": "model-00014-of-00021.safetensors", "model.layers.67.mlp.gate_proj.weight": "model-00014-of-00021.safetensors", "model.layers.87.mlp.up_proj.weight": "model-00014-of-00021.safetensors", "model.layers.67.mlp.up_proj.weight": "model-00014-of-00021.safetensors", "model.layers.87.post_attention_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.67.post_attention_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.87.input_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.67.input_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.86.self_attn.o_proj.weight": "model-00014-of-00021.safetensors", "model.layers.66.self_attn.o_proj.weight": "model-00014-of-00021.safetensors", "model.layers.86.self_attn.v_proj.weight": "model-00014-of-00021.safetensors", "model.layers.66.self_attn.v_proj.weight": "model-00014-of-00021.safetensors", "model.layers.86.self_attn.k_proj.weight": "model-00014-of-00021.safetensors", "model.layers.66.self_attn.k_proj.weight": "model-00014-of-00021.safetensors", "model.layers.86.self_attn.q_proj.weight": "model-00014-of-00021.safetensors", "model.layers.66.self_attn.q_proj.weight": "model-00014-of-00021.safetensors", "model.layers.86.input_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.66.input_layernorm.weight": "model-00014-of-00021.safetensors", "model.layers.85.mlp.down_proj.weight": "model-00014-of-00021.safetensors", "model.layers.65.mlp.down_proj.weight": "model-00014-of-00021.safetensors", "model.layers.85.mlp.gate_proj.weight": "model-00014-of-00021.safetensors", "model.layers.65.mlp.gate_proj.weight": "model-00014-of-00021.safetensors", "model.layers.85.mlp.up_proj.weight": "model-00015-of-00021.safetensors", "model.layers.65.mlp.up_proj.weight": "model-00015-of-00021.safetensors", "model.layers.85.post_attention_layernorm.weight": "model-00015-of-00021.safetensors", "model.layers.65.post_attention_layernorm.weight": "model-00015-of-00021.safetensors", "model.layers.85.self_attn.o_proj.weight": "model-00015-of-00021.safetensors", "model.layers.65.self_attn.o_proj.weight": "model-00015-of-00021.safetensors", "model.layers.85.self_attn.v_proj.weight": "model-00015-of-00021.safetensors", "model.layers.65.self_attn.v_proj.weight": "model-00015-of-00021.safetensors", "model.layers.85.self_attn.k_proj.weight": "model-00015-of-00021.safetensors", "model.layers.65.self_attn.k_proj.weight": "model-00015-of-00021.safetensors", "model.layers.85.self_attn.q_proj.weight": "model-00015-of-00021.safetensors", "model.layers.65.self_attn.q_proj.weight": "model-00015-of-00021.safetensors", "model.layers.85.input_layernorm.weight": "model-00015-of-00021.safetensors", "model.layers.65.input_layernorm.weight": "model-00015-of-00021.safetensors", "model.layers.84.mlp.down_proj.weight": "model-00015-of-00021.safetensors", "model.layers.64.mlp.down_proj.weight": "model-00015-of-00021.safetensors", "model.layers.84.mlp.up_proj.weight": "model-00015-of-00021.safetensors", "model.layers.64.mlp.up_proj.weight": "model-00015-of-00021.safetensors", "model.layers.84.post_attention_layernorm.weight": "model-00015-of-00021.safetensors", "model.layers.64.post_attention_layernorm.weight": "model-00015-of-00021.safetensors", "model.layers.84.input_layernorm.weight": "model-00015-of-00021.safetensors", "model.layers.64.input_layernorm.weight": "model-00015-of-00021.safetensors", "model.layers.98.mlp.gate_proj.weight": "model-00015-of-00021.safetensors", "model.layers.78.mlp.gate_proj.weight": "model-00015-of-00021.safetensors", "model.layers.98.self_attn.o_proj.weight": "model-00015-of-00021.safetensors", "model.layers.78.self_attn.o_proj.weight": "model-00015-of-00021.safetensors", "model.layers.98.self_attn.v_proj.weight": "model-00015-of-00021.safetensors", "model.layers.78.self_attn.v_proj.weight": "model-00015-of-00021.safetensors", "model.layers.98.self_attn.k_proj.weight": "model-00015-of-00021.safetensors", "model.layers.78.self_attn.k_proj.weight": "model-00015-of-00021.safetensors", "model.layers.98.self_attn.q_proj.weight": "model-00015-of-00021.safetensors", "model.layers.78.self_attn.q_proj.weight": "model-00015-of-00021.safetensors", "model.layers.97.mlp.down_proj.weight": "model-00015-of-00021.safetensors", "model.layers.77.mlp.down_proj.weight": "model-00015-of-00021.safetensors", "model.layers.97.mlp.gate_proj.weight": "model-00015-of-00021.safetensors", "model.layers.77.mlp.gate_proj.weight": "model-00015-of-00021.safetensors", "model.layers.95.mlp.gate_proj.weight": "model-00015-of-00021.safetensors", "model.layers.75.mlp.gate_proj.weight": "model-00015-of-00021.safetensors", "model.layers.95.mlp.up_proj.weight": "model-00015-of-00021.safetensors", "model.layers.75.mlp.up_proj.weight": "model-00015-of-00021.safetensors", "model.layers.95.self_attn.o_proj.weight": "model-00015-of-00021.safetensors", "model.layers.75.self_attn.o_proj.weight": "model-00015-of-00021.safetensors", "model.layers.95.self_attn.v_proj.weight": "model-00015-of-00021.safetensors", "model.layers.75.self_attn.v_proj.weight": "model-00015-of-00021.safetensors", "model.layers.95.self_attn.k_proj.weight": "model-00015-of-00021.safetensors", "model.layers.75.self_attn.k_proj.weight": "model-00015-of-00021.safetensors", "model.layers.95.self_attn.q_proj.weight": "model-00015-of-00021.safetensors", "model.layers.75.self_attn.q_proj.weight": "model-00015-of-00021.safetensors", "model.layers.94.mlp.down_proj.weight": "model-00015-of-00021.safetensors", "model.layers.74.mlp.down_proj.weight": "model-00016-of-00021.safetensors", "model.layers.94.mlp.gate_proj.weight": "model-00016-of-00021.safetensors", "model.layers.74.mlp.gate_proj.weight": "model-00016-of-00021.safetensors", "model.layers.94.mlp.up_proj.weight": "model-00016-of-00021.safetensors", "model.layers.74.mlp.up_proj.weight": "model-00016-of-00021.safetensors", "model.layers.94.post_attention_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.74.post_attention_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.94.self_attn.o_proj.weight": "model-00016-of-00021.safetensors", "model.layers.74.self_attn.o_proj.weight": "model-00016-of-00021.safetensors", "model.layers.94.self_attn.v_proj.weight": "model-00016-of-00021.safetensors", "model.layers.74.self_attn.v_proj.weight": "model-00016-of-00021.safetensors", "model.layers.94.self_attn.k_proj.weight": "model-00016-of-00021.safetensors", "model.layers.74.self_attn.k_proj.weight": "model-00016-of-00021.safetensors", "model.layers.94.self_attn.q_proj.weight": "model-00016-of-00021.safetensors", "model.layers.74.self_attn.q_proj.weight": "model-00016-of-00021.safetensors", "model.layers.94.input_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.74.input_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.93.mlp.down_proj.weight": "model-00016-of-00021.safetensors", "model.layers.73.mlp.down_proj.weight": "model-00016-of-00021.safetensors", "model.layers.93.mlp.gate_proj.weight": "model-00016-of-00021.safetensors", "model.layers.73.mlp.gate_proj.weight": "model-00016-of-00021.safetensors", "model.layers.93.mlp.up_proj.weight": "model-00016-of-00021.safetensors", "model.layers.73.mlp.up_proj.weight": "model-00016-of-00021.safetensors", "model.layers.93.post_attention_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.73.post_attention_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.93.self_attn.o_proj.weight": "model-00016-of-00021.safetensors", "model.layers.73.self_attn.o_proj.weight": "model-00016-of-00021.safetensors", "model.layers.93.self_attn.v_proj.weight": "model-00016-of-00021.safetensors", "model.layers.73.self_attn.v_proj.weight": "model-00016-of-00021.safetensors", "model.layers.93.self_attn.k_proj.weight": "model-00016-of-00021.safetensors", "model.layers.73.self_attn.k_proj.weight": "model-00016-of-00021.safetensors", "model.layers.93.self_attn.q_proj.weight": "model-00016-of-00021.safetensors", "model.layers.73.self_attn.q_proj.weight": "model-00016-of-00021.safetensors", "model.layers.93.input_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.73.input_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.91.mlp.down_proj.weight": "model-00016-of-00021.safetensors", "model.layers.71.mlp.down_proj.weight": "model-00016-of-00021.safetensors", "model.layers.91.mlp.gate_proj.weight": "model-00016-of-00021.safetensors", "model.layers.71.mlp.gate_proj.weight": "model-00016-of-00021.safetensors", "model.layers.91.mlp.up_proj.weight": "model-00016-of-00021.safetensors", "model.layers.71.mlp.up_proj.weight": "model-00016-of-00021.safetensors", "model.layers.91.post_attention_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.71.post_attention_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.91.self_attn.o_proj.weight": "model-00016-of-00021.safetensors", "model.layers.71.self_attn.o_proj.weight": "model-00016-of-00021.safetensors", "model.layers.91.self_attn.v_proj.weight": "model-00016-of-00021.safetensors", "model.layers.71.self_attn.v_proj.weight": "model-00016-of-00021.safetensors", "model.layers.91.self_attn.k_proj.weight": "model-00016-of-00021.safetensors", "model.layers.71.self_attn.k_proj.weight": "model-00016-of-00021.safetensors", "model.layers.91.self_attn.q_proj.weight": "model-00016-of-00021.safetensors", "model.layers.71.self_attn.q_proj.weight": "model-00016-of-00021.safetensors", "model.layers.91.input_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.71.input_layernorm.weight": "model-00016-of-00021.safetensors", "model.layers.90.mlp.down_proj.weight": "model-00017-of-00021.safetensors", "model.layers.70.mlp.down_proj.weight": "model-00017-of-00021.safetensors", "model.layers.90.mlp.gate_proj.weight": "model-00017-of-00021.safetensors", "model.layers.70.mlp.gate_proj.weight": "model-00017-of-00021.safetensors", "model.layers.90.mlp.up_proj.weight": "model-00017-of-00021.safetensors", "model.layers.70.mlp.up_proj.weight": "model-00017-of-00021.safetensors", "model.layers.90.post_attention_layernorm.weight": "model-00017-of-00021.safetensors", "model.layers.70.post_attention_layernorm.weight": "model-00017-of-00021.safetensors", "model.layers.90.self_attn.o_proj.weight": "model-00017-of-00021.safetensors", "model.layers.70.self_attn.o_proj.weight": "model-00017-of-00021.safetensors", "model.layers.90.input_layernorm.weight": "model-00017-of-00021.safetensors", "model.layers.70.input_layernorm.weight": "model-00017-of-00021.safetensors", "model.layers.104.self_attn.v_proj.weight": "model-00017-of-00021.safetensors", "model.layers.104.self_attn.k_proj.weight": "model-00017-of-00021.safetensors", "model.layers.104.self_attn.q_proj.weight": "model-00017-of-00021.safetensors", "model.layers.103.mlp.down_proj.weight": "model-00017-of-00021.safetensors", "model.layers.103.mlp.gate_proj.weight": "model-00017-of-00021.safetensors", "model.layers.103.mlp.up_proj.weight": "model-00017-of-00021.safetensors", "model.layers.103.post_attention_layernorm.weight": "model-00017-of-00021.safetensors", "model.layers.103.self_attn.o_proj.weight": "model-00017-of-00021.safetensors", "model.layers.103.self_attn.v_proj.weight": "model-00017-of-00021.safetensors", "model.layers.103.self_attn.k_proj.weight": "model-00017-of-00021.safetensors", "model.layers.103.self_attn.q_proj.weight": "model-00017-of-00021.safetensors", "model.layers.103.input_layernorm.weight": "model-00017-of-00021.safetensors", "model.layers.102.mlp.down_proj.weight": "model-00017-of-00021.safetensors", "model.layers.101.self_attn.o_proj.weight": "model-00017-of-00021.safetensors", "model.layers.101.self_attn.v_proj.weight": "model-00017-of-00021.safetensors", "model.layers.101.self_attn.k_proj.weight": "model-00017-of-00021.safetensors", "model.layers.101.self_attn.q_proj.weight": "model-00017-of-00021.safetensors", "model.layers.100.mlp.down_proj.weight": "model-00017-of-00021.safetensors", "model.layers.100.mlp.gate_proj.weight": "model-00017-of-00021.safetensors", "model.layers.100.mlp.up_proj.weight": "model-00017-of-00021.safetensors", "model.layers.100.post_attention_layernorm.weight": "model-00017-of-00021.safetensors", "model.layers.100.self_attn.o_proj.weight": "model-00017-of-00021.safetensors", "model.layers.100.self_attn.v_proj.weight": "model-00017-of-00021.safetensors", "model.layers.100.self_attn.k_proj.weight": "model-00017-of-00021.safetensors", "model.layers.100.self_attn.q_proj.weight": "model-00017-of-00021.safetensors", "model.layers.100.input_layernorm.weight": "model-00017-of-00021.safetensors", "model.layers.99.mlp.down_proj.weight": "model-00017-of-00021.safetensors", "model.layers.79.mlp.down_proj.weight": "model-00017-of-00021.safetensors", "model.layers.99.mlp.gate_proj.weight": "model-00017-of-00021.safetensors", "model.layers.79.mlp.gate_proj.weight": "model-00017-of-00021.safetensors", "model.layers.99.mlp.up_proj.weight": "model-00017-of-00021.safetensors", "model.layers.79.mlp.up_proj.weight": "model-00018-of-00021.safetensors", "model.layers.99.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.79.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.99.self_attn.o_proj.weight": "model-00018-of-00021.safetensors", "model.layers.79.self_attn.o_proj.weight": "model-00018-of-00021.safetensors", "model.layers.99.self_attn.v_proj.weight": "model-00018-of-00021.safetensors", "model.layers.79.self_attn.v_proj.weight": "model-00018-of-00021.safetensors", "model.layers.99.self_attn.k_proj.weight": "model-00018-of-00021.safetensors", "model.layers.79.self_attn.k_proj.weight": "model-00018-of-00021.safetensors", "model.layers.99.self_attn.q_proj.weight": "model-00018-of-00021.safetensors", "model.layers.79.self_attn.q_proj.weight": "model-00018-of-00021.safetensors", "model.layers.99.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.79.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.98.mlp.down_proj.weight": "model-00018-of-00021.safetensors", "model.layers.78.mlp.down_proj.weight": "model-00018-of-00021.safetensors", "model.layers.98.mlp.up_proj.weight": "model-00018-of-00021.safetensors", "model.layers.78.mlp.up_proj.weight": "model-00018-of-00021.safetensors", "model.layers.98.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.78.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.98.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.78.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.97.mlp.up_proj.weight": "model-00018-of-00021.safetensors", "model.layers.77.mlp.up_proj.weight": "model-00018-of-00021.safetensors", "model.layers.97.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.77.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.97.self_attn.o_proj.weight": "model-00018-of-00021.safetensors", "model.layers.77.self_attn.o_proj.weight": "model-00018-of-00021.safetensors", "model.layers.97.self_attn.v_proj.weight": "model-00018-of-00021.safetensors", "model.layers.77.self_attn.v_proj.weight": "model-00018-of-00021.safetensors", "model.layers.97.self_attn.k_proj.weight": "model-00018-of-00021.safetensors", "model.layers.77.self_attn.k_proj.weight": "model-00018-of-00021.safetensors", "model.layers.97.self_attn.q_proj.weight": "model-00018-of-00021.safetensors", "model.layers.77.self_attn.q_proj.weight": "model-00018-of-00021.safetensors", "model.layers.97.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.77.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.96.mlp.down_proj.weight": "model-00018-of-00021.safetensors", "model.layers.76.mlp.down_proj.weight": "model-00018-of-00021.safetensors", "model.layers.96.mlp.gate_proj.weight": "model-00018-of-00021.safetensors", "model.layers.76.mlp.gate_proj.weight": "model-00018-of-00021.safetensors", "model.layers.96.mlp.up_proj.weight": "model-00018-of-00021.safetensors", "model.layers.76.mlp.up_proj.weight": "model-00018-of-00021.safetensors", "model.layers.96.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.76.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.96.self_attn.o_proj.weight": "model-00018-of-00021.safetensors", "model.layers.76.self_attn.o_proj.weight": "model-00018-of-00021.safetensors", "model.layers.96.self_attn.v_proj.weight": "model-00018-of-00021.safetensors", "model.layers.76.self_attn.v_proj.weight": "model-00018-of-00021.safetensors", "model.layers.96.self_attn.k_proj.weight": "model-00018-of-00021.safetensors", "model.layers.76.self_attn.k_proj.weight": "model-00018-of-00021.safetensors", "model.layers.96.self_attn.q_proj.weight": "model-00018-of-00021.safetensors", "model.layers.76.self_attn.q_proj.weight": "model-00018-of-00021.safetensors", "model.layers.96.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.76.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.95.mlp.down_proj.weight": "model-00018-of-00021.safetensors", "model.layers.75.mlp.down_proj.weight": "model-00018-of-00021.safetensors", "model.layers.95.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.75.post_attention_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.95.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.75.input_layernorm.weight": "model-00018-of-00021.safetensors", "model.layers.109.mlp.gate_proj.weight": "model-00018-of-00021.safetensors", "model.layers.109.mlp.up_proj.weight": "model-00018-of-00021.safetensors", "model.layers.109.self_attn.o_proj.weight": "model-00018-of-00021.safetensors", "model.layers.109.self_attn.v_proj.weight": "model-00018-of-00021.safetensors", "model.layers.109.self_attn.k_proj.weight": "model-00018-of-00021.safetensors", "model.layers.109.self_attn.q_proj.weight": "model-00019-of-00021.safetensors", "model.layers.108.mlp.down_proj.weight": "model-00019-of-00021.safetensors", "model.layers.108.mlp.gate_proj.weight": "model-00019-of-00021.safetensors", "model.layers.106.mlp.down_proj.weight": "model-00019-of-00021.safetensors", "model.layers.106.mlp.gate_proj.weight": "model-00019-of-00021.safetensors", "model.layers.106.mlp.up_proj.weight": "model-00019-of-00021.safetensors", "model.layers.106.post_attention_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.106.self_attn.o_proj.weight": "model-00019-of-00021.safetensors", "model.layers.106.self_attn.v_proj.weight": "model-00019-of-00021.safetensors", "model.layers.106.self_attn.k_proj.weight": "model-00019-of-00021.safetensors", "model.layers.106.self_attn.q_proj.weight": "model-00019-of-00021.safetensors", "model.layers.106.input_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.105.mlp.down_proj.weight": "model-00019-of-00021.safetensors", "model.layers.105.mlp.gate_proj.weight": "model-00019-of-00021.safetensors", "model.layers.105.mlp.up_proj.weight": "model-00019-of-00021.safetensors", "model.layers.105.post_attention_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.105.self_attn.o_proj.weight": "model-00019-of-00021.safetensors", "model.layers.105.self_attn.v_proj.weight": "model-00019-of-00021.safetensors", "model.layers.105.self_attn.k_proj.weight": "model-00019-of-00021.safetensors", "model.layers.105.self_attn.q_proj.weight": "model-00019-of-00021.safetensors", "model.layers.105.input_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.104.mlp.down_proj.weight": "model-00019-of-00021.safetensors", "model.layers.104.mlp.gate_proj.weight": "model-00019-of-00021.safetensors", "model.layers.104.mlp.up_proj.weight": "model-00019-of-00021.safetensors", "model.layers.104.post_attention_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.104.self_attn.o_proj.weight": "model-00019-of-00021.safetensors", "model.layers.104.input_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.102.mlp.gate_proj.weight": "model-00019-of-00021.safetensors", "model.layers.102.mlp.up_proj.weight": "model-00019-of-00021.safetensors", "model.layers.102.post_attention_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.102.self_attn.o_proj.weight": "model-00019-of-00021.safetensors", "model.layers.102.self_attn.v_proj.weight": "model-00019-of-00021.safetensors", "model.layers.102.self_attn.k_proj.weight": "model-00019-of-00021.safetensors", "model.layers.102.self_attn.q_proj.weight": "model-00019-of-00021.safetensors", "model.layers.102.input_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.101.mlp.down_proj.weight": "model-00019-of-00021.safetensors", "model.layers.101.mlp.gate_proj.weight": "model-00019-of-00021.safetensors", "model.layers.101.mlp.up_proj.weight": "model-00019-of-00021.safetensors", "model.layers.101.post_attention_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.101.input_layernorm.weight": "model-00019-of-00021.safetensors", "model.layers.115.self_attn.o_proj.weight": "model-00019-of-00021.safetensors", "model.layers.115.self_attn.v_proj.weight": "model-00019-of-00021.safetensors", "model.layers.115.self_attn.k_proj.weight": "model-00019-of-00021.safetensors", "model.layers.115.self_attn.q_proj.weight": "model-00019-of-00021.safetensors", "model.layers.114.mlp.down_proj.weight": "model-00019-of-00021.safetensors", "model.layers.114.mlp.gate_proj.weight": "model-00019-of-00021.safetensors", "model.layers.114.mlp.up_proj.weight": "model-00020-of-00021.safetensors", "model.layers.114.post_attention_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.114.self_attn.o_proj.weight": "model-00020-of-00021.safetensors", "model.layers.114.self_attn.v_proj.weight": "model-00020-of-00021.safetensors", "model.layers.114.self_attn.k_proj.weight": "model-00020-of-00021.safetensors", "model.layers.114.self_attn.q_proj.weight": "model-00020-of-00021.safetensors", "model.layers.114.input_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.113.mlp.down_proj.weight": "model-00020-of-00021.safetensors", "model.layers.112.mlp.gate_proj.weight": "model-00020-of-00021.safetensors", "model.layers.112.self_attn.o_proj.weight": "model-00020-of-00021.safetensors", "model.layers.112.self_attn.v_proj.weight": "model-00020-of-00021.safetensors", "model.layers.112.self_attn.k_proj.weight": "model-00020-of-00021.safetensors", "model.layers.112.self_attn.q_proj.weight": "model-00020-of-00021.safetensors", "model.layers.111.mlp.down_proj.weight": "model-00020-of-00021.safetensors", "model.layers.111.mlp.gate_proj.weight": "model-00020-of-00021.safetensors", "model.layers.111.mlp.up_proj.weight": "model-00020-of-00021.safetensors", "model.layers.111.post_attention_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.111.self_attn.o_proj.weight": "model-00020-of-00021.safetensors", "model.layers.111.self_attn.v_proj.weight": "model-00020-of-00021.safetensors", "model.layers.111.self_attn.k_proj.weight": "model-00020-of-00021.safetensors", "model.layers.111.self_attn.q_proj.weight": "model-00020-of-00021.safetensors", "model.layers.111.input_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.110.mlp.down_proj.weight": "model-00020-of-00021.safetensors", "model.layers.110.mlp.gate_proj.weight": "model-00020-of-00021.safetensors", "model.layers.110.mlp.up_proj.weight": "model-00020-of-00021.safetensors", "model.layers.110.post_attention_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.110.self_attn.o_proj.weight": "model-00020-of-00021.safetensors", "model.layers.110.self_attn.v_proj.weight": "model-00020-of-00021.safetensors", "model.layers.110.self_attn.k_proj.weight": "model-00020-of-00021.safetensors", "model.layers.110.self_attn.q_proj.weight": "model-00020-of-00021.safetensors", "model.layers.110.input_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.109.mlp.down_proj.weight": "model-00020-of-00021.safetensors", "model.layers.109.post_attention_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.109.input_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.108.mlp.up_proj.weight": "model-00020-of-00021.safetensors", "model.layers.108.post_attention_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.108.self_attn.o_proj.weight": "model-00020-of-00021.safetensors", "model.layers.108.self_attn.v_proj.weight": "model-00020-of-00021.safetensors", "model.layers.108.self_attn.k_proj.weight": "model-00020-of-00021.safetensors", "model.layers.108.self_attn.q_proj.weight": "model-00020-of-00021.safetensors", "model.layers.108.input_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.107.mlp.down_proj.weight": "model-00020-of-00021.safetensors", "model.layers.107.mlp.gate_proj.weight": "model-00020-of-00021.safetensors", "model.layers.107.mlp.up_proj.weight": "model-00020-of-00021.safetensors", "model.layers.107.post_attention_layernorm.weight": "model-00020-of-00021.safetensors", "model.layers.107.self_attn.o_proj.weight": "model-00020-of-00021.safetensors", "model.layers.107.self_attn.v_proj.weight": "model-00020-of-00021.safetensors", "model.layers.107.self_attn.k_proj.weight": "model-00020-of-00021.safetensors", "model.layers.107.self_attn.q_proj.weight": "model-00020-of-00021.safetensors", "model.layers.107.input_layernorm.weight": "model-00020-of-00021.safetensors", "lm_head.weight": "model-00020-of-00021.safetensors", "model.norm.weight": "model-00020-of-00021.safetensors", "model.layers.119.mlp.down_proj.weight": "model-00020-of-00021.safetensors", "model.layers.119.mlp.gate_proj.weight": "model-00020-of-00021.safetensors", "model.layers.119.mlp.up_proj.weight": "model-00021-of-00021.safetensors", "model.layers.119.post_attention_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.119.self_attn.o_proj.weight": "model-00021-of-00021.safetensors", "model.layers.119.self_attn.v_proj.weight": "model-00021-of-00021.safetensors", "model.layers.119.self_attn.k_proj.weight": "model-00021-of-00021.safetensors", "model.layers.119.self_attn.q_proj.weight": "model-00021-of-00021.safetensors", "model.layers.119.input_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.118.mlp.down_proj.weight": "model-00021-of-00021.safetensors", "model.layers.118.self_attn.v_proj.weight": "model-00021-of-00021.safetensors", "model.layers.118.self_attn.k_proj.weight": "model-00021-of-00021.safetensors", "model.layers.118.self_attn.q_proj.weight": "model-00021-of-00021.safetensors", "model.layers.117.mlp.down_proj.weight": "model-00021-of-00021.safetensors", "model.layers.117.mlp.gate_proj.weight": "model-00021-of-00021.safetensors", "model.layers.117.mlp.up_proj.weight": "model-00021-of-00021.safetensors", "model.layers.117.post_attention_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.117.self_attn.o_proj.weight": "model-00021-of-00021.safetensors", "model.layers.117.self_attn.v_proj.weight": "model-00021-of-00021.safetensors", "model.layers.117.self_attn.k_proj.weight": "model-00021-of-00021.safetensors", "model.layers.117.self_attn.q_proj.weight": "model-00021-of-00021.safetensors", "model.layers.117.input_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.116.mlp.down_proj.weight": "model-00021-of-00021.safetensors", "model.layers.116.mlp.gate_proj.weight": "model-00021-of-00021.safetensors", "model.layers.116.mlp.up_proj.weight": "model-00021-of-00021.safetensors", "model.layers.116.post_attention_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.116.self_attn.o_proj.weight": "model-00021-of-00021.safetensors", "model.layers.116.self_attn.v_proj.weight": "model-00021-of-00021.safetensors", "model.layers.116.self_attn.k_proj.weight": "model-00021-of-00021.safetensors", "model.layers.116.self_attn.q_proj.weight": "model-00021-of-00021.safetensors", "model.layers.116.input_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.115.mlp.down_proj.weight": "model-00021-of-00021.safetensors", "model.layers.115.mlp.gate_proj.weight": "model-00021-of-00021.safetensors", "model.layers.115.mlp.up_proj.weight": "model-00021-of-00021.safetensors", "model.layers.115.post_attention_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.115.input_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.113.mlp.gate_proj.weight": "model-00021-of-00021.safetensors", "model.layers.113.mlp.up_proj.weight": "model-00021-of-00021.safetensors", "model.layers.113.post_attention_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.113.self_attn.o_proj.weight": "model-00021-of-00021.safetensors", "model.layers.113.self_attn.v_proj.weight": "model-00021-of-00021.safetensors", "model.layers.113.self_attn.k_proj.weight": "model-00021-of-00021.safetensors", "model.layers.113.self_attn.q_proj.weight": "model-00021-of-00021.safetensors", "model.layers.113.input_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.112.mlp.down_proj.weight": "model-00021-of-00021.safetensors", "model.layers.112.mlp.up_proj.weight": "model-00021-of-00021.safetensors", "model.layers.112.post_attention_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.112.input_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.118.mlp.gate_proj.weight": "model-00021-of-00021.safetensors", "model.layers.118.mlp.up_proj.weight": "model-00021-of-00021.safetensors", "model.layers.118.post_attention_layernorm.weight": "model-00021-of-00021.safetensors", "model.layers.118.self_attn.o_proj.weight": "model-00021-of-00021.safetensors", "model.layers.118.input_layernorm.weight": "model-00021-of-00021.safetensors"}}
output-00001-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4234c877f9fd36dfae1e5b0e8b80efecbb42de80868bb750a2b471ffaaa8d432
3
+ size 8482138934
output-00002-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d4ebb7e409f01cb20fab6739609c9338d8eb64e48004701bbf1c65f66c89e52
3
+ size 8556703768
output-00003-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b57b6233b99e485c74d06e3396877c32a9a4ce831e13fa19ea8a3cea1bf8a15c
3
+ size 8588032328
output-00004-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3ab191d4f87af24bf57cd0dacb6a7c561549c98d8f2130b603fb1faade9831d6
3
+ size 8588159368
output-00005-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68a752c328552707af80a43b09d11d310e4da6c59691bdfec71722f3eaa46afc
3
+ size 8577178672
output-00006-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3f0c65d747fded1d56069a611c95d60e56cd2e188106d54676a3a087cb81498
3
+ size 8549762808
output-00007-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2064afafa9d2b1acf7638d13b8d0db05c92884902003fe0f61740a16386efd6a
3
+ size 7147337752
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<unk>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": true,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": false,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": "<unk>",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false
42
+ }