lucid-gunner commited on
Commit
3c6672c
1 Parent(s): e075d2c

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +146 -0
README.md ADDED
@@ -0,0 +1,146 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: dnhkng/RYS-Llama-3.1-8B-Instruct
3
+ license: mit
4
+ tags:
5
+ - llama-cpp
6
+ - gguf-my-repo
7
+ model-index:
8
+ - name: RYS-Llama-3.1-8B-Instruct
9
+ results:
10
+ - task:
11
+ type: text-generation
12
+ name: Text Generation
13
+ dataset:
14
+ name: IFEval (0-Shot)
15
+ type: HuggingFaceH4/ifeval
16
+ args:
17
+ num_few_shot: 0
18
+ metrics:
19
+ - type: inst_level_strict_acc and prompt_level_strict_acc
20
+ value: 76.85
21
+ name: strict accuracy
22
+ source:
23
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=dnhkng/RYS-Llama-3.1-8B-Instruct
24
+ name: Open LLM Leaderboard
25
+ - task:
26
+ type: text-generation
27
+ name: Text Generation
28
+ dataset:
29
+ name: BBH (3-Shot)
30
+ type: BBH
31
+ args:
32
+ num_few_shot: 3
33
+ metrics:
34
+ - type: acc_norm
35
+ value: 31.09
36
+ name: normalized accuracy
37
+ source:
38
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=dnhkng/RYS-Llama-3.1-8B-Instruct
39
+ name: Open LLM Leaderboard
40
+ - task:
41
+ type: text-generation
42
+ name: Text Generation
43
+ dataset:
44
+ name: MATH Lvl 5 (4-Shot)
45
+ type: hendrycks/competition_math
46
+ args:
47
+ num_few_shot: 4
48
+ metrics:
49
+ - type: exact_match
50
+ value: 11.33
51
+ name: exact match
52
+ source:
53
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=dnhkng/RYS-Llama-3.1-8B-Instruct
54
+ name: Open LLM Leaderboard
55
+ - task:
56
+ type: text-generation
57
+ name: Text Generation
58
+ dataset:
59
+ name: GPQA (0-shot)
60
+ type: Idavidrein/gpqa
61
+ args:
62
+ num_few_shot: 0
63
+ metrics:
64
+ - type: acc_norm
65
+ value: 2.35
66
+ name: acc_norm
67
+ source:
68
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=dnhkng/RYS-Llama-3.1-8B-Instruct
69
+ name: Open LLM Leaderboard
70
+ - task:
71
+ type: text-generation
72
+ name: Text Generation
73
+ dataset:
74
+ name: MuSR (0-shot)
75
+ type: TAUR-Lab/MuSR
76
+ args:
77
+ num_few_shot: 0
78
+ metrics:
79
+ - type: acc_norm
80
+ value: 7.68
81
+ name: acc_norm
82
+ source:
83
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=dnhkng/RYS-Llama-3.1-8B-Instruct
84
+ name: Open LLM Leaderboard
85
+ - task:
86
+ type: text-generation
87
+ name: Text Generation
88
+ dataset:
89
+ name: MMLU-PRO (5-shot)
90
+ type: TIGER-Lab/MMLU-Pro
91
+ config: main
92
+ split: test
93
+ args:
94
+ num_few_shot: 5
95
+ metrics:
96
+ - type: acc
97
+ value: 29.33
98
+ name: accuracy
99
+ source:
100
+ url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=dnhkng/RYS-Llama-3.1-8B-Instruct
101
+ name: Open LLM Leaderboard
102
+ ---
103
+
104
+ # lucid-gunner/RYS-Llama-3.1-8B-Instruct-Q5_K_M-GGUF
105
+ This model was converted to GGUF format from [`dnhkng/RYS-Llama-3.1-8B-Instruct`](https://huggingface.co/dnhkng/RYS-Llama-3.1-8B-Instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
106
+ Refer to the [original model card](https://huggingface.co/dnhkng/RYS-Llama-3.1-8B-Instruct) for more details on the model.
107
+
108
+ ## Use with llama.cpp
109
+ Install llama.cpp through brew (works on Mac and Linux)
110
+
111
+ ```bash
112
+ brew install llama.cpp
113
+
114
+ ```
115
+ Invoke the llama.cpp server or the CLI.
116
+
117
+ ### CLI:
118
+ ```bash
119
+ llama-cli --hf-repo lucid-gunner/RYS-Llama-3.1-8B-Instruct-Q5_K_M-GGUF --hf-file rys-llama-3.1-8b-instruct-q5_k_m.gguf -p "The meaning to life and the universe is"
120
+ ```
121
+
122
+ ### Server:
123
+ ```bash
124
+ llama-server --hf-repo lucid-gunner/RYS-Llama-3.1-8B-Instruct-Q5_K_M-GGUF --hf-file rys-llama-3.1-8b-instruct-q5_k_m.gguf -c 2048
125
+ ```
126
+
127
+ 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.
128
+
129
+ Step 1: Clone llama.cpp from GitHub.
130
+ ```
131
+ git clone https://github.com/ggerganov/llama.cpp
132
+ ```
133
+
134
+ Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
135
+ ```
136
+ cd llama.cpp && LLAMA_CURL=1 make
137
+ ```
138
+
139
+ Step 3: Run inference through the main binary.
140
+ ```
141
+ ./llama-cli --hf-repo lucid-gunner/RYS-Llama-3.1-8B-Instruct-Q5_K_M-GGUF --hf-file rys-llama-3.1-8b-instruct-q5_k_m.gguf -p "The meaning to life and the universe is"
142
+ ```
143
+ or
144
+ ```
145
+ ./llama-server --hf-repo lucid-gunner/RYS-Llama-3.1-8B-Instruct-Q5_K_M-GGUF --hf-file rys-llama-3.1-8b-instruct-q5_k_m.gguf -c 2048
146
+ ```