Hugging Face's logo

itlwas
/
Sailor-0.5B-Q4_K_M-GGUF

GGUF
multilingual
sea
sailor
llama-cpp
gguf-my-repo
Eval Results (legacy)
conversational
Model card Files
xet
 Community
itlwas commited on
Commit
4b8326e
·
verified ·
1 Parent(s): 37637fa

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +174 -0
README.md ADDED
@@ -0,0 +1,174 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ - zh
5
+ - id
6
+ - th
7
+ - vi
8
+ - ms
9
+ - lo
10
+ license: apache-2.0
11
+ tags:
12
+ - multilingual
13
+ - sea
14
+ - sailor
15
+ - llama-cpp
16
+ - gguf-my-repo
17
+ base_model: Qwen/Qwen1.5-0.5B
18
+ datasets:
19
+ - cerebras/SlimPajama-627B
20
+ - Skywork/SkyPile-150B
21
+ - allenai/MADLAD-400
22
+ - cc100
23
+ inference: false
24
+ model-index:
25
+ - name: Sailor-0.5B
26
+ results:
27
+ - task:
28
+ type: text-generation
29
+ dataset:
30
+ name: XQuAD-Thai
31
+ type: XQuAD-Thai
32
+ metrics:
33
+ - type: EM (3-Shot)
34
+ value: 15.84
35
+ name: EM (3-Shot)
36
+ - type: F1 (3-Shot)
37
+ value: 27.58
38
+ name: F1 (3-Shot)
39
+ - task:
40
+ type: text-generation
41
+ dataset:
42
+ name: TyDiQA-Indonesian
43
+ type: TyDiQA-Indonesian
44
+ metrics:
45
+ - type: EM (3-Shot)
46
+ value: 30.44
47
+ name: EM (3-Shot)
48
+ - type: F1 (3-Shot)
49
+ value: 54.74
50
+ name: F1 (3-Shot)
51
+ - task:
52
+ type: text-generation
53
+ dataset:
54
+ name: XQuAD-Vietnamese
55
+ type: XQuAD-Vietnamese
56
+ metrics:
57
+ - type: EM (3-Shot)
58
+ value: 21.13
59
+ name: EM (3-Shot)
60
+ - type: F1 (3-Shot)
61
+ value: 40.57
62
+ name: F1 (3-Shot)
63
+ - task:
64
+ type: text-generation
65
+ dataset:
66
+ name: XCOPA-Thai
67
+ type: XCOPA-Thai
68
+ metrics:
69
+ - type: EM (3-Shot)
70
+ value: 51.0
71
+ name: EM (3-Shot)
72
+ - task:
73
+ type: text-generation
74
+ dataset:
75
+ name: XCOPA-Indonesian
76
+ type: XCOPA-Indonesian
77
+ metrics:
78
+ - type: EM (3-Shot)
79
+ value: 58.2
80
+ name: EM (3-Shot)
81
+ - task:
82
+ type: text-generation
83
+ dataset:
84
+ name: XCOPA-Vietnamese
85
+ type: XCOPA-Vietnamese
86
+ metrics:
87
+ - type: EM (3-Shot)
88
+ value: 58.0
89
+ name: EM (3-Shot)
90
+ - task:
91
+ type: text-generation
92
+ dataset:
93
+ name: M3Exam-Thai
94
+ type: M3Exam-Thai
95
+ metrics:
96
+ - type: EM (3-Shot)
97
+ value: 24.41
98
+ name: EM (3-Shot)
99
+ - task:
100
+ type: text-generation
101
+ dataset:
102
+ name: M3Exam-Indonesian
103
+ type: M3Exam-Indonesian
104
+ metrics:
105
+ - type: EM (3-Shot)
106
+ value: 26.15
107
+ name: EM (3-Shot)
108
+ - task:
109
+ type: text-generation
110
+ dataset:
111
+ name: M3Exam-Vietnamese
112
+ type: M3Exam-Vietnamese
113
+ metrics:
114
+ - type: EM (3-Shot)
115
+ value: 30.91
116
+ name: EM (3-Shot)
117
+ - task:
118
+ type: text-generation
119
+ dataset:
120
+ name: BELEBELE-Thai
121
+ type: BELEBELE-Thai
122
+ metrics:
123
+ - type: EM (3-Shot)
124
+ value: 32.22
125
+ name: EM (3-Shot)
126
+ - task:
127
+ type: text-generation
128
+ dataset:
129
+ name: BELEBELE-Indonesian
130
+ type: BELEBELE-Indonesian
131
+ metrics:
132
+ - type: EM (3-Shot)
133
+ value: 30.89
134
+ name: EM (3-Shot)
135
+ - task:
136
+ type: text-generation
137
+ dataset:
138
+ name: BELEBELE-Vietnamese
139
+ type: BELEBELE-Vietnamese
140
+ metrics:
141
+ - type: EM (3-Shot)
142
+ value: 32.33
143
+ name: EM (3-Shot)
144
+ ---
145
+
146
+ # AIronMind/Sailor-0.5B-Q4_K_M-GGUF
147
+ This model was converted to GGUF format from [`sail/Sailor-0.5B`](https://huggingface.co/sail/Sailor-0.5B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
148
+ Refer to the [original model card](https://huggingface.co/sail/Sailor-0.5B) for more details on the model.
149
+ ## Use with llama.cpp
150
+
151
+ Install llama.cpp through brew.
152
+
153
+ ```bash
154
+ brew install ggerganov/ggerganov/llama.cpp
155
+ ```
156
+ Invoke the llama.cpp server or the CLI.
157
+
158
+ CLI:
159
+
160
+ ```bash
161
+ llama-cli --hf-repo AIronMind/Sailor-0.5B-Q4_K_M-GGUF --model sailor-0.5b.Q4_K_M.gguf -p "The meaning to life and the universe is"
162
+ ```
163
+
164
+ Server:
165
+
166
+ ```bash
167
+ llama-server --hf-repo AIronMind/Sailor-0.5B-Q4_K_M-GGUF --model sailor-0.5b.Q4_K_M.gguf -c 2048
168
+ ```
169
+
170
+ Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
171
+
172
+ ```
173
+ git clone https://github.com/ggerganov/llama.cpp && cd llama.cpp && make && ./main -m sailor-0.5b.Q4_K_M.gguf -n 128
174
+ ```