Hugging Face's logo

Nexesenex
/
Mistral-7B-Instruct-v0.2-2x7B-MoE-6.0bpw-h6-exl2

Text Generation
Transformers
Safetensors
mixtral
finetuned
conversational
text-generation-inference
Model card Files Community
Nexesenex commited on
Commit
6664b8b
·
1 Parent(s): 0ddde1a

Upload 12 files

Browse files
Files changed (12) hide show
  1. README.md +84 -0
  2. config.json +30 -0
  3. generation_config.json +6 -0
  4. huggingface-metadata.txt +11 -0
  5. model.safetensors.index.json +426 -0
  6. output-00001-of-00002.safetensors +3 -0
  7. output-00002-of-00002.safetensors +3 -0
  8. perlthoughts_Mistral-7B-Instruct-v0.2-2x7B-MoE_measurement.json +0 -0
  9. special_tokens_map.json +30 -0
  10. tokenizer.json +0 -0
  11. tokenizer.model +3 -0
  12. tokenizer_config.json +50 -0
README.md CHANGED
@@ -1,3 +1,87 @@
1
  ---
2
  license: apache-2.0
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
+ pipeline_tag: text-generation
4
+ tags:
5
+ - finetuned
6
+ inference: false
7
  ---
8
+
9
+ Mistral 7B Instruct v0.2 7B (with only 2 experts)
10
+
11
+ # Model Card for Mistral-7B-Instruct-v0.2
12
+
13
+ The Mistral-7B-Instruct-v0.2 Large Language Model (LLM) is an improved instruct fine-tuned version of [Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1).
14
+
15
+ For full details of this model please read our [paper](https://arxiv.org/abs/2310.06825) and [release blog post](https://mistral.ai/news/la-plateforme/).
16
+
17
+ ## Instruction format
18
+
19
+ In order to leverage instruction fine-tuning, your prompt should be surrounded by `[INST]` and `[/INST]` tokens. The very first instruction should begin with a begin of sentence id. The next instructions should not. The assistant generation will be ended by the end-of-sentence token id.
20
+
21
+ E.g.
22
+ ```
23
+ text = "<s>[INST] What is your favourite condiment? [/INST]"
24
+ "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!</s> "
25
+ "[INST] Do you have mayonnaise recipes? [/INST]"
26
+ ```
27
+
28
+ This format is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating) via the `apply_chat_template()` method:
29
+
30
+ ```python
31
+ from transformers import AutoModelForCausalLM, AutoTokenizer
32
+
33
+ device = "cuda" # the device to load the model onto
34
+
35
+ model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
36
+ tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
37
+
38
+ messages = [
39
+ {"role": "user", "content": "What is your favourite condiment?"},
40
+ {"role": "assistant", "content": "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!"},
41
+ {"role": "user", "content": "Do you have mayonnaise recipes?"}
42
+ ]
43
+
44
+ encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")
45
+
46
+ model_inputs = encodeds.to(device)
47
+ model.to(device)
48
+
49
+ generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
50
+ decoded = tokenizer.batch_decode(generated_ids)
51
+ print(decoded[0])
52
+ ```
53
+
54
+ ## Model Architecture
55
+ This instruction model is based on Mistral-7B-v0.1, a transformer model with the following architecture choices:
56
+ - Grouped-Query Attention
57
+ - Sliding-Window Attention
58
+ - Byte-fallback BPE tokenizer
59
+
60
+ ## Troubleshooting
61
+ - If you see the following error:
62
+ ```
63
+ Traceback (most recent call last):
64
+ File "", line 1, in
65
+ File "/transformers/models/auto/auto_factory.py", line 482, in from_pretrained
66
+ config, kwargs = AutoConfig.from_pretrained(
67
+ File "/transformers/models/auto/configuration_auto.py", line 1022, in from_pretrained
68
+ config_class = CONFIG_MAPPING[config_dict["model_type"]]
69
+ File "/transformers/models/auto/configuration_auto.py", line 723, in getitem
70
+ raise KeyError(key)
71
+ KeyError: 'mistral'
72
+ ```
73
+
74
+ Installing transformers from source should solve the issue
75
+ pip install git+https://github.com/huggingface/transformers
76
+
77
+ This should not be required after transformers-v4.33.4.
78
+
79
+ ## Limitations
80
+
81
+ The Mistral 7B Instruct model is a quick demonstration that the base model can be easily fine-tuned to achieve compelling performance.
82
+ It does not have any moderation mechanisms. We're looking forward to engaging with the community on ways to
83
+ make the model finely respect guardrails, allowing for deployment in environments requiring moderated outputs.
84
+
85
+ ## The Mistral AI Team
86
+
87
+ Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Blanche Savary, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Emma Bou Hanna, Florian Bressand, Gianna Lengyel, Guillaume Bour, Guillaume Lample, Lélio Renard Lavaud, Louis Ternon, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Théophile Gervet, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/workspace/new-merged",
3
+ "architectures": [
4
+ "MixtralForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 4096,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 14336,
13
+ "max_position_embeddings": 32768,
14
+ "model_type": "mixtral",
15
+ "num_attention_heads": 32,
16
+ "num_experts_per_tok": 2,
17
+ "num_hidden_layers": 32,
18
+ "num_key_value_heads": 8,
19
+ "num_local_experts": 2,
20
+ "output_router_logits": false,
21
+ "rms_norm_eps": 1e-05,
22
+ "rope_theta": 1000000.0,
23
+ "router_aux_loss_coef": 0.001,
24
+ "sliding_window": null,
25
+ "tie_word_embeddings": false,
26
+ "torch_dtype": "float16",
27
+ "transformers_version": "4.37.0.dev0",
28
+ "use_cache": true,
29
+ "vocab_size": 32000
30
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.37.0.dev0"
6
+ }
huggingface-metadata.txt ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ url: https://huggingface.co/perlthoughts/Mistral-7B-Instruct-v0.2-2x7B-MoE
2
+ branch: main
3
+ download date: 2023-12-30 09:55:24
4
+ sha256sum:
5
+ 36323c861f8953c93fc7fcabad74a457c0bcd97a27904ed2f0062f3a17fe4e51 model-00001-of-00006.safetensors
6
+ b4297bf03de898f9673b9078793f5f46bfcfeb4c0b54314fd3c40d99a24e4e6a model-00002-of-00006.safetensors
7
+ bd95a5e425cbd1b383d4c6d1077824b5350840f244dd8b78a80949b3bc7defe4 model-00003-of-00006.safetensors
8
+ 1c23da870e4000be1dedf2b1748587788e18ee859ceae091d9d8035fa6a3fcdd model-00004-of-00006.safetensors
9
+ cdf6f72055823d3b72bb5126ffe492ac516718413ea418861c1af4642c346aa9 model-00005-of-00006.safetensors
10
+ c5a64ea31960822bfed0e201abb07be63f2cffee165d5d44f920d717fb01effb model-00006-of-00006.safetensors
11
+ dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055 tokenizer.model
model.safetensors.index.json ADDED
@@ -0,0 +1,426 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 25758277632
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00006-of-00006.safetensors",
7
+ "model.embed_tokens.weight": "model-00001-of-00006.safetensors",
8
+ "model.layers.0.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00006.safetensors",
9
+ "model.layers.0.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00006.safetensors",
10
+ "model.layers.0.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00006.safetensors",
11
+ "model.layers.0.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00006.safetensors",
12
+ "model.layers.0.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00006.safetensors",
13
+ "model.layers.0.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00006.safetensors",
14
+ "model.layers.0.block_sparse_moe.gate.weight": "model-00001-of-00006.safetensors",
15
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00006.safetensors",
16
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
17
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
18
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
19
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
20
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
21
+ "model.layers.1.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00006.safetensors",
22
+ "model.layers.1.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00006.safetensors",
23
+ "model.layers.1.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00006.safetensors",
24
+ "model.layers.1.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00006.safetensors",
25
+ "model.layers.1.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00006.safetensors",
26
+ "model.layers.1.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00006.safetensors",
27
+ "model.layers.1.block_sparse_moe.gate.weight": "model-00001-of-00006.safetensors",
28
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00006.safetensors",
29
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
30
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
31
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
32
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
33
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
34
+ "model.layers.10.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00006.safetensors",
35
+ "model.layers.10.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00006.safetensors",
36
+ "model.layers.10.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00006.safetensors",
37
+ "model.layers.10.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00006.safetensors",
38
+ "model.layers.10.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00006.safetensors",
39
+ "model.layers.10.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00006.safetensors",
40
+ "model.layers.10.block_sparse_moe.gate.weight": "model-00002-of-00006.safetensors",
41
+ "model.layers.10.input_layernorm.weight": "model-00002-of-00006.safetensors",
42
+ "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
43
+ "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
44
+ "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
45
+ "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
46
+ "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
47
+ "model.layers.11.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00006.safetensors",
48
+ "model.layers.11.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00006.safetensors",
49
+ "model.layers.11.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00006.safetensors",
50
+ "model.layers.11.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00006.safetensors",
51
+ "model.layers.11.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00006.safetensors",
52
+ "model.layers.11.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00006.safetensors",
53
+ "model.layers.11.block_sparse_moe.gate.weight": "model-00002-of-00006.safetensors",
54
+ "model.layers.11.input_layernorm.weight": "model-00002-of-00006.safetensors",
55
+ "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
56
+ "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
57
+ "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
58
+ "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
59
+ "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
60
+ "model.layers.12.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00006.safetensors",
61
+ "model.layers.12.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00006.safetensors",
62
+ "model.layers.12.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00006.safetensors",
63
+ "model.layers.12.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00006.safetensors",
64
+ "model.layers.12.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00006.safetensors",
65
+ "model.layers.12.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00006.safetensors",
66
+ "model.layers.12.block_sparse_moe.gate.weight": "model-00002-of-00006.safetensors",
67
+ "model.layers.12.input_layernorm.weight": "model-00003-of-00006.safetensors",
68
+ "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
69
+ "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
70
+ "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
71
+ "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
72
+ "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
73
+ "model.layers.13.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00006.safetensors",
74
+ "model.layers.13.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00006.safetensors",
75
+ "model.layers.13.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00006.safetensors",
76
+ "model.layers.13.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00006.safetensors",
77
+ "model.layers.13.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00006.safetensors",
78
+ "model.layers.13.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00006.safetensors",
79
+ "model.layers.13.block_sparse_moe.gate.weight": "model-00003-of-00006.safetensors",
80
+ "model.layers.13.input_layernorm.weight": "model-00003-of-00006.safetensors",
81
+ "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
82
+ "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
83
+ "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
84
+ "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
85
+ "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
86
+ "model.layers.14.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00006.safetensors",
87
+ "model.layers.14.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00006.safetensors",
88
+ "model.layers.14.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00006.safetensors",
89
+ "model.layers.14.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00006.safetensors",
90
+ "model.layers.14.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00006.safetensors",
91
+ "model.layers.14.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00006.safetensors",
92
+ "model.layers.14.block_sparse_moe.gate.weight": "model-00003-of-00006.safetensors",
93
+ "model.layers.14.input_layernorm.weight": "model-00003-of-00006.safetensors",
94
+ "model.layers.14.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
95
+ "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
96
+ "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
97
+ "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
98
+ "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
99
+ "model.layers.15.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00006.safetensors",
100
+ "model.layers.15.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00006.safetensors",
101
+ "model.layers.15.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00006.safetensors",
102
+ "model.layers.15.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00006.safetensors",
103
+ "model.layers.15.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00006.safetensors",
104
+ "model.layers.15.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00006.safetensors",
105
+ "model.layers.15.block_sparse_moe.gate.weight": "model-00003-of-00006.safetensors",
106
+ "model.layers.15.input_layernorm.weight": "model-00003-of-00006.safetensors",
107
+ "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
108
+ "model.layers.15.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
109
+ "model.layers.15.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
110
+ "model.layers.15.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
111
+ "model.layers.15.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
112
+ "model.layers.16.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00006.safetensors",
113
+ "model.layers.16.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00006.safetensors",
114
+ "model.layers.16.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00006.safetensors",
115
+ "model.layers.16.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00006.safetensors",
116
+ "model.layers.16.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00006.safetensors",
117
+ "model.layers.16.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00006.safetensors",
118
+ "model.layers.16.block_sparse_moe.gate.weight": "model-00003-of-00006.safetensors",
119
+ "model.layers.16.input_layernorm.weight": "model-00003-of-00006.safetensors",
120
+ "model.layers.16.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
121
+ "model.layers.16.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
122
+ "model.layers.16.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
123
+ "model.layers.16.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
124
+ "model.layers.16.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
125
+ "model.layers.17.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00006.safetensors",
126
+ "model.layers.17.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00006.safetensors",
127
+ "model.layers.17.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00006.safetensors",
128
+ "model.layers.17.block_sparse_moe.experts.1.w1.weight": "model-00003-of-00006.safetensors",
129
+ "model.layers.17.block_sparse_moe.experts.1.w2.weight": "model-00003-of-00006.safetensors",
130
+ "model.layers.17.block_sparse_moe.experts.1.w3.weight": "model-00003-of-00006.safetensors",
131
+ "model.layers.17.block_sparse_moe.gate.weight": "model-00003-of-00006.safetensors",
132
+ "model.layers.17.input_layernorm.weight": "model-00003-of-00006.safetensors",
133
+ "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
134
+ "model.layers.17.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
135
+ "model.layers.17.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
136
+ "model.layers.17.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
137
+ "model.layers.17.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
138
+ "model.layers.18.block_sparse_moe.experts.0.w1.weight": "model-00003-of-00006.safetensors",
139
+ "model.layers.18.block_sparse_moe.experts.0.w2.weight": "model-00003-of-00006.safetensors",
140
+ "model.layers.18.block_sparse_moe.experts.0.w3.weight": "model-00003-of-00006.safetensors",
141
+ "model.layers.18.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00006.safetensors",
142
+ "model.layers.18.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00006.safetensors",
143
+ "model.layers.18.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00006.safetensors",
144
+ "model.layers.18.block_sparse_moe.gate.weight": "model-00003-of-00006.safetensors",
145
+ "model.layers.18.input_layernorm.weight": "model-00004-of-00006.safetensors",
146
+ "model.layers.18.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
147
+ "model.layers.18.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
148
+ "model.layers.18.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
149
+ "model.layers.18.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
150
+ "model.layers.18.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
151
+ "model.layers.19.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00006.safetensors",
152
+ "model.layers.19.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00006.safetensors",
153
+ "model.layers.19.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00006.safetensors",
154
+ "model.layers.19.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00006.safetensors",
155
+ "model.layers.19.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00006.safetensors",
156
+ "model.layers.19.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00006.safetensors",
157
+ "model.layers.19.block_sparse_moe.gate.weight": "model-00004-of-00006.safetensors",
158
+ "model.layers.19.input_layernorm.weight": "model-00004-of-00006.safetensors",
159
+ "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
160
+ "model.layers.19.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
161
+ "model.layers.19.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
162
+ "model.layers.19.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
163
+ "model.layers.19.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
164
+ "model.layers.2.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00006.safetensors",
165
+ "model.layers.2.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00006.safetensors",
166
+ "model.layers.2.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00006.safetensors",
167
+ "model.layers.2.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00006.safetensors",
168
+ "model.layers.2.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00006.safetensors",
169
+ "model.layers.2.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00006.safetensors",
170
+ "model.layers.2.block_sparse_moe.gate.weight": "model-00001-of-00006.safetensors",
171
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00006.safetensors",
172
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
173
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
174
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
175
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
176
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
177
+ "model.layers.20.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00006.safetensors",
178
+ "model.layers.20.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00006.safetensors",
179
+ "model.layers.20.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00006.safetensors",
180
+ "model.layers.20.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00006.safetensors",
181
+ "model.layers.20.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00006.safetensors",
182
+ "model.layers.20.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00006.safetensors",
183
+ "model.layers.20.block_sparse_moe.gate.weight": "model-00004-of-00006.safetensors",
184
+ "model.layers.20.input_layernorm.weight": "model-00004-of-00006.safetensors",
185
+ "model.layers.20.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
186
+ "model.layers.20.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
187
+ "model.layers.20.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
188
+ "model.layers.20.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
189
+ "model.layers.20.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
190
+ "model.layers.21.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00006.safetensors",
191
+ "model.layers.21.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00006.safetensors",
192
+ "model.layers.21.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00006.safetensors",
193
+ "model.layers.21.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00006.safetensors",
194
+ "model.layers.21.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00006.safetensors",
195
+ "model.layers.21.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00006.safetensors",
196
+ "model.layers.21.block_sparse_moe.gate.weight": "model-00004-of-00006.safetensors",
197
+ "model.layers.21.input_layernorm.weight": "model-00004-of-00006.safetensors",
198
+ "model.layers.21.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
199
+ "model.layers.21.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
200
+ "model.layers.21.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
201
+ "model.layers.21.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
202
+ "model.layers.21.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
203
+ "model.layers.22.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00006.safetensors",
204
+ "model.layers.22.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00006.safetensors",
205
+ "model.layers.22.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00006.safetensors",
206
+ "model.layers.22.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00006.safetensors",
207
+ "model.layers.22.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00006.safetensors",
208
+ "model.layers.22.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00006.safetensors",
209
+ "model.layers.22.block_sparse_moe.gate.weight": "model-00004-of-00006.safetensors",
210
+ "model.layers.22.input_layernorm.weight": "model-00004-of-00006.safetensors",
211
+ "model.layers.22.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
212
+ "model.layers.22.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
213
+ "model.layers.22.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
214
+ "model.layers.22.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
215
+ "model.layers.22.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
216
+ "model.layers.23.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00006.safetensors",
217
+ "model.layers.23.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00006.safetensors",
218
+ "model.layers.23.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00006.safetensors",
219
+ "model.layers.23.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00006.safetensors",
220
+ "model.layers.23.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00006.safetensors",
221
+ "model.layers.23.block_sparse_moe.experts.1.w3.weight": "model-00004-of-00006.safetensors",
222
+ "model.layers.23.block_sparse_moe.gate.weight": "model-00004-of-00006.safetensors",
223
+ "model.layers.23.input_layernorm.weight": "model-00004-of-00006.safetensors",
224
+ "model.layers.23.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
225
+ "model.layers.23.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
226
+ "model.layers.23.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
227
+ "model.layers.23.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
228
+ "model.layers.23.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
229
+ "model.layers.24.block_sparse_moe.experts.0.w1.weight": "model-00004-of-00006.safetensors",
230
+ "model.layers.24.block_sparse_moe.experts.0.w2.weight": "model-00004-of-00006.safetensors",
231
+ "model.layers.24.block_sparse_moe.experts.0.w3.weight": "model-00004-of-00006.safetensors",
232
+ "model.layers.24.block_sparse_moe.experts.1.w1.weight": "model-00004-of-00006.safetensors",
233
+ "model.layers.24.block_sparse_moe.experts.1.w2.weight": "model-00004-of-00006.safetensors",
234
+ "model.layers.24.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00006.safetensors",
235
+ "model.layers.24.block_sparse_moe.gate.weight": "model-00004-of-00006.safetensors",
236
+ "model.layers.24.input_layernorm.weight": "model-00005-of-00006.safetensors",
237
+ "model.layers.24.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
238
+ "model.layers.24.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
239
+ "model.layers.24.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
240
+ "model.layers.24.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
241
+ "model.layers.24.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
242
+ "model.layers.25.block_sparse_moe.experts.0.w1.weight": "model-00005-of-00006.safetensors",
243
+ "model.layers.25.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00006.safetensors",
244
+ "model.layers.25.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00006.safetensors",
245
+ "model.layers.25.block_sparse_moe.experts.1.w1.weight": "model-00005-of-00006.safetensors",
246
+ "model.layers.25.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00006.safetensors",
247
+ "model.layers.25.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00006.safetensors",
248
+ "model.layers.25.block_sparse_moe.gate.weight": "model-00005-of-00006.safetensors",
249
+ "model.layers.25.input_layernorm.weight": "model-00005-of-00006.safetensors",
250
+ "model.layers.25.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
251
+ "model.layers.25.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
252
+ "model.layers.25.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
253
+ "model.layers.25.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
254
+ "model.layers.25.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
255
+ "model.layers.26.block_sparse_moe.experts.0.w1.weight": "model-00005-of-00006.safetensors",
256
+ "model.layers.26.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00006.safetensors",
257
+ "model.layers.26.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00006.safetensors",
258
+ "model.layers.26.block_sparse_moe.experts.1.w1.weight": "model-00005-of-00006.safetensors",
259
+ "model.layers.26.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00006.safetensors",
260
+ "model.layers.26.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00006.safetensors",
261
+ "model.layers.26.block_sparse_moe.gate.weight": "model-00005-of-00006.safetensors",
262
+ "model.layers.26.input_layernorm.weight": "model-00005-of-00006.safetensors",
263
+ "model.layers.26.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
264
+ "model.layers.26.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
265
+ "model.layers.26.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
266
+ "model.layers.26.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
267
+ "model.layers.26.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
268
+ "model.layers.27.block_sparse_moe.experts.0.w1.weight": "model-00005-of-00006.safetensors",
269
+ "model.layers.27.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00006.safetensors",
270
+ "model.layers.27.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00006.safetensors",
271
+ "model.layers.27.block_sparse_moe.experts.1.w1.weight": "model-00005-of-00006.safetensors",
272
+ "model.layers.27.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00006.safetensors",
273
+ "model.layers.27.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00006.safetensors",
274
+ "model.layers.27.block_sparse_moe.gate.weight": "model-00005-of-00006.safetensors",
275
+ "model.layers.27.input_layernorm.weight": "model-00005-of-00006.safetensors",
276
+ "model.layers.27.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
277
+ "model.layers.27.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
278
+ "model.layers.27.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
279
+ "model.layers.27.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
280
+ "model.layers.27.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
281
+ "model.layers.28.block_sparse_moe.experts.0.w1.weight": "model-00005-of-00006.safetensors",
282
+ "model.layers.28.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00006.safetensors",
283
+ "model.layers.28.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00006.safetensors",
284
+ "model.layers.28.block_sparse_moe.experts.1.w1.weight": "model-00005-of-00006.safetensors",
285
+ "model.layers.28.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00006.safetensors",
286
+ "model.layers.28.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00006.safetensors",
287
+ "model.layers.28.block_sparse_moe.gate.weight": "model-00005-of-00006.safetensors",
288
+ "model.layers.28.input_layernorm.weight": "model-00005-of-00006.safetensors",
289
+ "model.layers.28.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
290
+ "model.layers.28.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
291
+ "model.layers.28.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
292
+ "model.layers.28.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
293
+ "model.layers.28.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
294
+ "model.layers.29.block_sparse_moe.experts.0.w1.weight": "model-00005-of-00006.safetensors",
295
+ "model.layers.29.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00006.safetensors",
296
+ "model.layers.29.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00006.safetensors",
297
+ "model.layers.29.block_sparse_moe.experts.1.w1.weight": "model-00005-of-00006.safetensors",
298
+ "model.layers.29.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00006.safetensors",
299
+ "model.layers.29.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00006.safetensors",
300
+ "model.layers.29.block_sparse_moe.gate.weight": "model-00005-of-00006.safetensors",
301
+ "model.layers.29.input_layernorm.weight": "model-00005-of-00006.safetensors",
302
+ "model.layers.29.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
303
+ "model.layers.29.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
304
+ "model.layers.29.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
305
+ "model.layers.29.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
306
+ "model.layers.29.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
307
+ "model.layers.3.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00006.safetensors",
308
+ "model.layers.3.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00006.safetensors",
309
+ "model.layers.3.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00006.safetensors",
310
+ "model.layers.3.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00006.safetensors",
311
+ "model.layers.3.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00006.safetensors",
312
+ "model.layers.3.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00006.safetensors",
313
+ "model.layers.3.block_sparse_moe.gate.weight": "model-00001-of-00006.safetensors",
314
+ "model.layers.3.input_layernorm.weight": "model-00001-of-00006.safetensors",
315
+ "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
316
+ "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
317
+ "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
318
+ "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
319
+ "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
320
+ "model.layers.30.block_sparse_moe.experts.0.w1.weight": "model-00005-of-00006.safetensors",
321
+ "model.layers.30.block_sparse_moe.experts.0.w2.weight": "model-00005-of-00006.safetensors",
322
+ "model.layers.30.block_sparse_moe.experts.0.w3.weight": "model-00005-of-00006.safetensors",
323
+ "model.layers.30.block_sparse_moe.experts.1.w1.weight": "model-00005-of-00006.safetensors",
324
+ "model.layers.30.block_sparse_moe.experts.1.w2.weight": "model-00005-of-00006.safetensors",
325
+ "model.layers.30.block_sparse_moe.experts.1.w3.weight": "model-00005-of-00006.safetensors",
326
+ "model.layers.30.block_sparse_moe.gate.weight": "model-00005-of-00006.safetensors",
327
+ "model.layers.30.input_layernorm.weight": "model-00005-of-00006.safetensors",
328
+ "model.layers.30.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
329
+ "model.layers.30.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
330
+ "model.layers.30.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
331
+ "model.layers.30.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
332
+ "model.layers.30.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
333
+ "model.layers.31.block_sparse_moe.experts.0.w1.weight": "model-00006-of-00006.safetensors",
334
+ "model.layers.31.block_sparse_moe.experts.0.w2.weight": "model-00006-of-00006.safetensors",
335
+ "model.layers.31.block_sparse_moe.experts.0.w3.weight": "model-00006-of-00006.safetensors",
336
+ "model.layers.31.block_sparse_moe.experts.1.w1.weight": "model-00006-of-00006.safetensors",
337
+ "model.layers.31.block_sparse_moe.experts.1.w2.weight": "model-00006-of-00006.safetensors",
338
+ "model.layers.31.block_sparse_moe.experts.1.w3.weight": "model-00006-of-00006.safetensors",
339
+ "model.layers.31.block_sparse_moe.gate.weight": "model-00005-of-00006.safetensors",
340
+ "model.layers.31.input_layernorm.weight": "model-00006-of-00006.safetensors",
341
+ "model.layers.31.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
342
+ "model.layers.31.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
343
+ "model.layers.31.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
344
+ "model.layers.31.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
345
+ "model.layers.31.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
346
+ "model.layers.4.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00006.safetensors",
347
+ "model.layers.4.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00006.safetensors",
348
+ "model.layers.4.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00006.safetensors",
349
+ "model.layers.4.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00006.safetensors",
350
+ "model.layers.4.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00006.safetensors",
351
+ "model.layers.4.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00006.safetensors",
352
+ "model.layers.4.block_sparse_moe.gate.weight": "model-00001-of-00006.safetensors",
353
+ "model.layers.4.input_layernorm.weight": "model-00001-of-00006.safetensors",
354
+ "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
355
+ "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
356
+ "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
357
+ "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
358
+ "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
359
+ "model.layers.5.block_sparse_moe.experts.0.w1.weight": "model-00001-of-00006.safetensors",
360
+ "model.layers.5.block_sparse_moe.experts.0.w2.weight": "model-00001-of-00006.safetensors",
361
+ "model.layers.5.block_sparse_moe.experts.0.w3.weight": "model-00001-of-00006.safetensors",
362
+ "model.layers.5.block_sparse_moe.experts.1.w1.weight": "model-00001-of-00006.safetensors",
363
+ "model.layers.5.block_sparse_moe.experts.1.w2.weight": "model-00001-of-00006.safetensors",
364
+ "model.layers.5.block_sparse_moe.experts.1.w3.weight": "model-00001-of-00006.safetensors",
365
+ "model.layers.5.block_sparse_moe.gate.weight": "model-00001-of-00006.safetensors",
366
+ "model.layers.5.input_layernorm.weight": "model-00001-of-00006.safetensors",
367
+ "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
368
+ "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
369
+ "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
370
+ "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
371
+ "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
372
+ "model.layers.6.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00006.safetensors",
373
+ "model.layers.6.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00006.safetensors",
374
+ "model.layers.6.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00006.safetensors",
375
+ "model.layers.6.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00006.safetensors",
376
+ "model.layers.6.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00006.safetensors",
377
+ "model.layers.6.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00006.safetensors",
378
+ "model.layers.6.block_sparse_moe.gate.weight": "model-00002-of-00006.safetensors",
379
+ "model.layers.6.input_layernorm.weight": "model-00002-of-00006.safetensors",
380
+ "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
381
+ "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
382
+ "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
383
+ "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
384
+ "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
385
+ "model.layers.7.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00006.safetensors",
386
+ "model.layers.7.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00006.safetensors",
387
+ "model.layers.7.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00006.safetensors",
388
+ "model.layers.7.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00006.safetensors",
389
+ "model.layers.7.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00006.safetensors",
390
+ "model.layers.7.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00006.safetensors",
391
+ "model.layers.7.block_sparse_moe.gate.weight": "model-00002-of-00006.safetensors",
392
+ "model.layers.7.input_layernorm.weight": "model-00002-of-00006.safetensors",
393
+ "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
394
+ "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
395
+ "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
396
+ "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
397
+ "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
398
+ "model.layers.8.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00006.safetensors",
399
+ "model.layers.8.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00006.safetensors",
400
+ "model.layers.8.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00006.safetensors",
401
+ "model.layers.8.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00006.safetensors",
402
+ "model.layers.8.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00006.safetensors",
403
+ "model.layers.8.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00006.safetensors",
404
+ "model.layers.8.block_sparse_moe.gate.weight": "model-00002-of-00006.safetensors",
405
+ "model.layers.8.input_layernorm.weight": "model-00002-of-00006.safetensors",
406
+ "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
407
+ "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
408
+ "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
409
+ "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
410
+ "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
411
+ "model.layers.9.block_sparse_moe.experts.0.w1.weight": "model-00002-of-00006.safetensors",
412
+ "model.layers.9.block_sparse_moe.experts.0.w2.weight": "model-00002-of-00006.safetensors",
413
+ "model.layers.9.block_sparse_moe.experts.0.w3.weight": "model-00002-of-00006.safetensors",
414
+ "model.layers.9.block_sparse_moe.experts.1.w1.weight": "model-00002-of-00006.safetensors",
415
+ "model.layers.9.block_sparse_moe.experts.1.w2.weight": "model-00002-of-00006.safetensors",
416
+ "model.layers.9.block_sparse_moe.experts.1.w3.weight": "model-00002-of-00006.safetensors",
417
+ "model.layers.9.block_sparse_moe.gate.weight": "model-00002-of-00006.safetensors",
418
+ "model.layers.9.input_layernorm.weight": "model-00002-of-00006.safetensors",
419
+ "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
420
+ "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
421
+ "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
422
+ "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
423
+ "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
424
+ "model.norm.weight": "model-00006-of-00006.safetensors"
425
+ }
426
+ }
output-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cb6805e437c6b1f8581bdbe09000f282b5562d511c968517654b397a9116765
3
+ size 8552526224
output-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfba87f5b3918685d79a0684254de4650252f1697a6b3cf6f578c15f3df26077
3
+ size 1283588448
perlthoughts_Mistral-7B-Instruct-v0.2-2x7B-MoE_measurement.json ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": false,
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:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [],
31
+ "bos_token": "<s>",
32
+ "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 %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "device_map": {
35
+ "": "cuda"
36
+ },
37
+ "eos_token": "</s>",
38
+ "legacy": true,
39
+ "max_length": null,
40
+ "model_max_length": 1000000000000000019884624838656,
41
+ "pad_to_multiple_of": null,
42
+ "pad_token": "<s>",
43
+ "pad_token_type_id": 0,
44
+ "padding_side": "left",
45
+ "sp_model_kwargs": {},
46
+ "spaces_between_special_tokens": false,
47
+ "tokenizer_class": "LlamaTokenizer",
48
+ "unk_token": "<unk>",
49
+ "use_default_system_prompt": false
50
+ }