Sri-Vigneshwar-DJ commited on
Commit
e8f5388
·
verified ·
1 Parent(s): e42fd45
Files changed (1) hide show
  1. README.md +70 -0
README.md ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: AIDC-AI/Marco-o1
3
+ library_name: transformers
4
+ license: other
5
+ tags:
6
+ - llama-cpp
7
+ - Marco-o1
8
+ - Marco
9
+ - o1
10
+ - GGUF
11
+ datasets: hawky_market_research_prompts
12
+ ---
13
+
14
+ # TODO:
15
+ 1. Datalake from https://github.com/ashwath007/HTGEN-ads-automation
16
+ 2. Funtuning the Detailed and Summary
17
+ 3. Data Set : https://github.com/ashwath007/HTGEN-ads-automation/ads/data/train
18
+ 4. Data Set(tail) : https://github.com/ashwath007/HTGEN-ads-automation/ads/data/train/tail
19
+
20
+ # Sri-Vigneshwar-DJ/sarvam-2b-v0.5-GGUF
21
+ This model was converted to GGUF format from [`AIDC-AI/Marco-o1`](https://huggingface.co/AIDC-AI/Marco-o1) using llama.cpp
22
+ Refer to the [original model card](https://huggingface.co/AIDC-AI/Marco-o1) for more details on the model.
23
+
24
+ ## Use with llama.cpp
25
+ Install llama.cpp through brew (works on Mac and Linux) from []
26
+
27
+ ```bash
28
+ brew install llama.cpp or !git clone https://github.com/ggerganov/llama.cpp.git
29
+
30
+ ```
31
+ Invoke the llama.cpp server or the CLI.
32
+
33
+ ### CLI:
34
+ ```bash
35
+ ! /content/llama.cpp/llama-cli -m ./Marco-o1-GGUF -n 90 --repeat_penalty 1.0 --color -i -r "User:" -f /content/llama.cpp/prompts/chat-with-bob.txt
36
+
37
+ or
38
+
39
+ llama-cli --hf-repo Sri-Vigneshwar-DJ/Marco-o1-GGUF --hf-file FP8.gguf -p "Create Meta Ads Templates"
40
+ ```
41
+
42
+ ### Server:
43
+ ```bash
44
+ llama-server --hf-repo Sri-Vigneshwar-DJ/Marco-o1-GGUF --hf-file FP8.gguf -c 2048
45
+ ```
46
+
47
+ 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.
48
+
49
+ Step 1: Clone llama.cpp from GitHub.
50
+ ```
51
+ git clone https://github.com/ggerganov/llama.cpp
52
+ ```
53
+
54
+ Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag or ''!make GGML_OPENBLAS=1' along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
55
+ ```
56
+ cd llama.cpp && LLAMA_CURL=1 make
57
+
58
+ or
59
+
60
+ !make GGML_OPENBLAS=1
61
+ ```
62
+
63
+ Step 3: Run inference through the main binary.
64
+ ```
65
+ ./llama-cli --hf-repo Sri-Vigneshwar-DJ/Marco-o1-GGUF --hf-file FP8.gguf -p "The meaning to life and the universe is"
66
+ ```
67
+ or
68
+ ```
69
+ ./llama-server --hf-repo Sri-Vigneshwar-DJ/Marco-o1-GGUF --hf-file sFP8.gguf -c 2048
70
+ ```