Commit
•
c03e302
1
Parent(s):
db8dc61
Upload README.md with huggingface_hub
Browse files
README.md
ADDED
@@ -0,0 +1,139 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: KOCDIGITAL/Kocdigital-LLM-8b-v0.1
|
3 |
+
language:
|
4 |
+
- tr
|
5 |
+
license: llama3
|
6 |
+
tags:
|
7 |
+
- llama-cpp
|
8 |
+
- gguf-my-repo
|
9 |
+
model-index:
|
10 |
+
- name: Kocdigital-LLM-8b-v0.1
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
type: text-generation
|
14 |
+
name: Text Generation
|
15 |
+
dataset:
|
16 |
+
name: AI2 Reasoning Challenge TR
|
17 |
+
type: ai2_arc
|
18 |
+
config: ARC-Challenge
|
19 |
+
split: test
|
20 |
+
args:
|
21 |
+
num_few_shot: 25
|
22 |
+
metrics:
|
23 |
+
- type: acc
|
24 |
+
value: 44.03
|
25 |
+
name: accuracy
|
26 |
+
- task:
|
27 |
+
type: text-generation
|
28 |
+
name: Text Generation
|
29 |
+
dataset:
|
30 |
+
name: HellaSwag TR
|
31 |
+
type: hellaswag
|
32 |
+
split: validation
|
33 |
+
args:
|
34 |
+
num_few_shot: 10
|
35 |
+
metrics:
|
36 |
+
- type: acc
|
37 |
+
value: 46.73
|
38 |
+
name: accuracy
|
39 |
+
- task:
|
40 |
+
type: text-generation
|
41 |
+
name: Text Generation
|
42 |
+
dataset:
|
43 |
+
name: MMLU TR
|
44 |
+
type: cais/mmlu
|
45 |
+
config: all
|
46 |
+
split: test
|
47 |
+
args:
|
48 |
+
num_few_shot: 5
|
49 |
+
metrics:
|
50 |
+
- type: acc
|
51 |
+
value: 49.11
|
52 |
+
name: accuracy
|
53 |
+
- task:
|
54 |
+
type: text-generation
|
55 |
+
name: Text Generation
|
56 |
+
dataset:
|
57 |
+
name: TruthfulQA TR
|
58 |
+
type: truthful_qa
|
59 |
+
config: multiple_choice
|
60 |
+
split: validation
|
61 |
+
args:
|
62 |
+
num_few_shot: 0
|
63 |
+
metrics:
|
64 |
+
- type: acc
|
65 |
+
value: 48.21
|
66 |
+
name: accuracy
|
67 |
+
- task:
|
68 |
+
type: text-generation
|
69 |
+
name: Text Generation
|
70 |
+
dataset:
|
71 |
+
name: Winogrande TR
|
72 |
+
type: winogrande
|
73 |
+
config: winogrande_xl
|
74 |
+
split: validation
|
75 |
+
args:
|
76 |
+
num_few_shot: 10
|
77 |
+
metrics:
|
78 |
+
- type: acc
|
79 |
+
value: 54.98
|
80 |
+
name: accuracy
|
81 |
+
- task:
|
82 |
+
type: text-generation
|
83 |
+
name: Text Generation
|
84 |
+
dataset:
|
85 |
+
name: GSM8k TR
|
86 |
+
type: gsm8k
|
87 |
+
config: main
|
88 |
+
split: test
|
89 |
+
args:
|
90 |
+
num_few_shot: 5
|
91 |
+
metrics:
|
92 |
+
- type: acc
|
93 |
+
value: 51.78
|
94 |
+
name: accuracy
|
95 |
+
---
|
96 |
+
|
97 |
+
# serkandyck/Kocdigital-LLM-8b-v0.1-Q4_K_M-GGUF
|
98 |
+
This model was converted to GGUF format from [`KOCDIGITAL/Kocdigital-LLM-8b-v0.1`](https://huggingface.co/KOCDIGITAL/Kocdigital-LLM-8b-v0.1) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
|
99 |
+
Refer to the [original model card](https://huggingface.co/KOCDIGITAL/Kocdigital-LLM-8b-v0.1) for more details on the model.
|
100 |
+
|
101 |
+
## Use with llama.cpp
|
102 |
+
Install llama.cpp through brew (works on Mac and Linux)
|
103 |
+
|
104 |
+
```bash
|
105 |
+
brew install llama.cpp
|
106 |
+
|
107 |
+
```
|
108 |
+
Invoke the llama.cpp server or the CLI.
|
109 |
+
|
110 |
+
### CLI:
|
111 |
+
```bash
|
112 |
+
llama-cli --hf-repo serkandyck/Kocdigital-LLM-8b-v0.1-Q4_K_M-GGUF --hf-file kocdigital-llm-8b-v0.1-q4_k_m.gguf -p "The meaning to life and the universe is"
|
113 |
+
```
|
114 |
+
|
115 |
+
### Server:
|
116 |
+
```bash
|
117 |
+
llama-server --hf-repo serkandyck/Kocdigital-LLM-8b-v0.1-Q4_K_M-GGUF --hf-file kocdigital-llm-8b-v0.1-q4_k_m.gguf -c 2048
|
118 |
+
```
|
119 |
+
|
120 |
+
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.
|
121 |
+
|
122 |
+
Step 1: Clone llama.cpp from GitHub.
|
123 |
+
```
|
124 |
+
git clone https://github.com/ggerganov/llama.cpp
|
125 |
+
```
|
126 |
+
|
127 |
+
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).
|
128 |
+
```
|
129 |
+
cd llama.cpp && LLAMA_CURL=1 make
|
130 |
+
```
|
131 |
+
|
132 |
+
Step 3: Run inference through the main binary.
|
133 |
+
```
|
134 |
+
./llama-cli --hf-repo serkandyck/Kocdigital-LLM-8b-v0.1-Q4_K_M-GGUF --hf-file kocdigital-llm-8b-v0.1-q4_k_m.gguf -p "The meaning to life and the universe is"
|
135 |
+
```
|
136 |
+
or
|
137 |
+
```
|
138 |
+
./llama-server --hf-repo serkandyck/Kocdigital-LLM-8b-v0.1-Q4_K_M-GGUF --hf-file kocdigital-llm-8b-v0.1-q4_k_m.gguf -c 2048
|
139 |
+
```
|