nold commited on
Commit
47d969a
1 Parent(s): e8c5b9d

Upload folder using huggingface_hub (#1)

Browse files

- 6c3ae9936a1e7251ae37f2f27b93981ca191b25c9d024592d48dbd984c94d8c6 (862fbc1a17d64f8c47653717ee06fd91428aa7c1)
- 8d01a80cb685f86446a6e015c1b60920d9a1089fd58a8085b48fbe7d58f32b68 (6156b777ea751aca6fc7006016a0160367ac886e)
- dfabdbced20890dd90356e01e17c037e432bf5bc57fd1b41f689ecfab6ca6cf2 (d6da1f79848cb356b31f0ae11719bb6c6f5ac7f1)
- 1931afd00344ebab065356dbd51fa36312b316e34858fa35c4f977dbdb2d67e4 (678fdc854d4373105b30dfa6efcf5da4ba3e3852)
- cba5fa4e633e9d8443996d7faaeb1dbd1bf5011d2aec39928e197340eb8b8286 (833088774e498f436ac14e2e5d9ab282169a86c8)
- e9ca6ebd0d555778088e9fa1c7c0645e40f43e339249d7cceed02fd0b6132adb (6cf5e5cdf025e3dc9116cc67bd991026b161857a)

.gitattributes CHANGED
@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ ChatHercules-2.5-Mistral-7B_Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
37
+ ChatHercules-2.5-Mistral-7B_Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
38
+ ChatHercules-2.5-Mistral-7B_Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
39
+ ChatHercules-2.5-Mistral-7B_Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
40
+ ChatHercules-2.5-Mistral-7B_Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
ChatHercules-2.5-Mistral-7B_Q3_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6917215fcdd2be50b9c9957013689f540ac6e9456f9ebc8d001f2efdff520f2
3
+ size 3518985984
ChatHercules-2.5-Mistral-7B_Q4_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ffb1267ab8c3ad19326b1d3f2e99f2f2a95470a04bd19d436a626e9ce3b0723
3
+ size 4368439040
ChatHercules-2.5-Mistral-7B_Q5_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fc91cbb6a55c837a0d2548c755e40fe7cb4071113a7b57ec4d6da3db095e634
3
+ size 5131409152
ChatHercules-2.5-Mistral-7B_Q6_K.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b0dd41a2bdc3aa967f5f4e793766f2f4cdf322512bebdbe518ed33c38016549
3
+ size 5942064896
ChatHercules-2.5-Mistral-7B_Q8_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c5f0ee77797cba19ea61147e50baa647ac53e7286f46340c09dddbf6ac7a4db
3
+ size 7695857408
README.md ADDED
@@ -0,0 +1,191 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - merge
5
+ - mergekit
6
+ - lazymergekit
7
+ - Locutusque/Hercules-2.5-Mistral-7B
8
+ - openchat/openchat-3.5-0106
9
+ base_model:
10
+ - Locutusque/Hercules-2.5-Mistral-7B
11
+ - openchat/openchat-3.5-0106
12
+ model-index:
13
+ - name: ChatHercules-2.5-Mistral-7B
14
+ results:
15
+ - task:
16
+ type: text-generation
17
+ name: Text Generation
18
+ dataset:
19
+ name: AI2 Reasoning Challenge (25-Shot)
20
+ type: ai2_arc
21
+ config: ARC-Challenge
22
+ split: test
23
+ args:
24
+ num_few_shot: 25
25
+ metrics:
26
+ - type: acc_norm
27
+ value: 65.1
28
+ name: normalized accuracy
29
+ source:
30
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=hydra-project/ChatHercules-2.5-Mistral-7B
31
+ name: Open LLM Leaderboard
32
+ - task:
33
+ type: text-generation
34
+ name: Text Generation
35
+ dataset:
36
+ name: HellaSwag (10-Shot)
37
+ type: hellaswag
38
+ split: validation
39
+ args:
40
+ num_few_shot: 10
41
+ metrics:
42
+ - type: acc_norm
43
+ value: 84.61
44
+ name: normalized accuracy
45
+ source:
46
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=hydra-project/ChatHercules-2.5-Mistral-7B
47
+ name: Open LLM Leaderboard
48
+ - task:
49
+ type: text-generation
50
+ name: Text Generation
51
+ dataset:
52
+ name: MMLU (5-Shot)
53
+ type: cais/mmlu
54
+ config: all
55
+ split: test
56
+ args:
57
+ num_few_shot: 5
58
+ metrics:
59
+ - type: acc
60
+ value: 65.35
61
+ name: accuracy
62
+ source:
63
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=hydra-project/ChatHercules-2.5-Mistral-7B
64
+ name: Open LLM Leaderboard
65
+ - task:
66
+ type: text-generation
67
+ name: Text Generation
68
+ dataset:
69
+ name: TruthfulQA (0-shot)
70
+ type: truthful_qa
71
+ config: multiple_choice
72
+ split: validation
73
+ args:
74
+ num_few_shot: 0
75
+ metrics:
76
+ - type: mc2
77
+ value: 47.52
78
+ source:
79
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=hydra-project/ChatHercules-2.5-Mistral-7B
80
+ name: Open LLM Leaderboard
81
+ - task:
82
+ type: text-generation
83
+ name: Text Generation
84
+ dataset:
85
+ name: Winogrande (5-shot)
86
+ type: winogrande
87
+ config: winogrande_xl
88
+ split: validation
89
+ args:
90
+ num_few_shot: 5
91
+ metrics:
92
+ - type: acc
93
+ value: 81.85
94
+ name: accuracy
95
+ source:
96
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=hydra-project/ChatHercules-2.5-Mistral-7B
97
+ name: Open LLM Leaderboard
98
+ - task:
99
+ type: text-generation
100
+ name: Text Generation
101
+ dataset:
102
+ name: GSM8k (5-shot)
103
+ type: gsm8k
104
+ config: main
105
+ split: test
106
+ args:
107
+ num_few_shot: 5
108
+ metrics:
109
+ - type: acc
110
+ value: 64.97
111
+ name: accuracy
112
+ source:
113
+ url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=hydra-project/ChatHercules-2.5-Mistral-7B
114
+ name: Open LLM Leaderboard
115
+ ---
116
+
117
+ # ChatHercules-2.5-Mistral-7B
118
+
119
+
120
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6437292ecd93f4c9a34b0d47/VW32vrPx2giqo5Od8Tyz0.png)
121
+
122
+ ChatHercules-2.5-Mistral-7B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
123
+ * [Locutusque/Hercules-2.5-Mistral-7B](https://huggingface.co/Locutusque/Hercules-2.5-Mistral-7B)
124
+ * [openchat/openchat-3.5-0106](https://huggingface.co/openchat/openchat-3.5-0106)
125
+
126
+
127
+
128
+ ## 🧩 Configuration
129
+
130
+ ```yaml
131
+ slices:
132
+ - sources:
133
+ - model: Locutusque/Hercules-2.5-Mistral-7B
134
+ layer_range: [0, 32]
135
+ - model: openchat/openchat-3.5-0106
136
+ layer_range: [0, 32]
137
+ merge_method: slerp
138
+ base_model: Locutusque/Hercules-2.5-Mistral-7B
139
+ parameters:
140
+ t:
141
+ - filter: self_attn
142
+ value: [0, 0.5, 0.3, 0.7, 1]
143
+ - filter: mlp
144
+ value: [1, 0.5, 0.7, 0.3, 0]
145
+ - value: 0.5
146
+ dtype: bfloat16
147
+ ```
148
+
149
+ ## 💻 Usage
150
+
151
+ ```python
152
+ !pip install -qU transformers accelerate
153
+
154
+ from transformers import AutoTokenizer
155
+ import transformers
156
+ import torch
157
+
158
+ model = "hydra-project/ChatHercules-2.5-Mistral-7B"
159
+ messages = [{"role": "user", "content": "What is a large language model?"}]
160
+
161
+ tokenizer = AutoTokenizer.from_pretrained(model)
162
+ prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
163
+ pipeline = transformers.pipeline(
164
+ "text-generation",
165
+ model=model,
166
+ torch_dtype=torch.float16,
167
+ device_map="auto",
168
+ )
169
+
170
+ outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
171
+ print(outputs[0]["generated_text"])
172
+ ```
173
+ # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
174
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_hydra-project__ChatHercules-2.5-Mistral-7B)
175
+
176
+ | Metric |Value|
177
+ |---------------------------------|----:|
178
+ |Avg. |68.24|
179
+ |AI2 Reasoning Challenge (25-Shot)|65.10|
180
+ |HellaSwag (10-Shot) |84.61|
181
+ |MMLU (5-Shot) |65.35|
182
+ |TruthfulQA (0-shot) |47.52|
183
+ |Winogrande (5-shot) |81.85|
184
+ |GSM8k (5-shot) |64.97|
185
+
186
+
187
+
188
+ ***
189
+
190
+ Quantization of Model [hydra-project/ChatHercules-2.5-Mistral-7B](https://huggingface.co/hydra-project/ChatHercules-2.5-Mistral-7B).
191
+ Created using [llm-quantizer](https://github.com/Nold360/llm-quantizer) Pipeline