Upload README.md with huggingface_hub
Browse files
README.md
ADDED
@@ -0,0 +1,98 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: Nekochu/Llama-2-13B-German-ORPO
|
3 |
+
datasets:
|
4 |
+
- mayflowergmbh/intel_orca_dpo_pairs_de
|
5 |
+
- LeoLM/OpenSchnabeltier
|
6 |
+
- LeoLM/German_Songs
|
7 |
+
- LeoLM/German_Poems
|
8 |
+
- bjoernp/ultrachat_de
|
9 |
+
- mayflowergmbh/ultra-chat_de
|
10 |
+
- mayflowergmbh/airoboros-3.0_de
|
11 |
+
- mayflowergmbh/booksum_de
|
12 |
+
- mayflowergmbh/dolphin_de
|
13 |
+
- mayflowergmbh/evol-instruct_de
|
14 |
+
- mayflowergmbh/openschnabeltier_de
|
15 |
+
- mayflowergmbh/alpaca-gpt4_de
|
16 |
+
- mayflowergmbh/dolly-15k_de
|
17 |
+
- mayflowergmbh/oasst_de
|
18 |
+
language:
|
19 |
+
- de
|
20 |
+
- en
|
21 |
+
library_name: peft
|
22 |
+
license: apache-2.0
|
23 |
+
pipeline_tag: text-generation
|
24 |
+
tags:
|
25 |
+
- llama-factory
|
26 |
+
- lora
|
27 |
+
- generated_from_trainer
|
28 |
+
- llama2
|
29 |
+
- llama
|
30 |
+
- instruct
|
31 |
+
- finetune
|
32 |
+
- llm
|
33 |
+
- pytorch
|
34 |
+
- llama-2
|
35 |
+
- german
|
36 |
+
- deutsch
|
37 |
+
- llama-cpp
|
38 |
+
- gguf-my-repo
|
39 |
+
model_creator: Nekochu
|
40 |
+
quantized_by: Nekochu
|
41 |
+
pretty_name: Llama-2 13B German ORPO
|
42 |
+
model_type: llama2
|
43 |
+
prompt_template: 'Below is an instruction that describes a task. Write a response
|
44 |
+
that appropriately completes the request. ### Instruction: {Instruction} {summary}
|
45 |
+
### input: {category} ### Response: {prompt}'
|
46 |
+
task_categories:
|
47 |
+
- question-answering
|
48 |
+
- text2text-generation
|
49 |
+
- conversational
|
50 |
+
inference: true
|
51 |
+
model-index:
|
52 |
+
- name: Llama-2-13B-German-ORPO
|
53 |
+
results: []
|
54 |
+
---
|
55 |
+
|
56 |
+
# jott1970/Llama-2-13B-German-ORPO-Q4_K_M-GGUF
|
57 |
+
This model was converted to GGUF format from [`Nekochu/Llama-2-13B-German-ORPO`](https://huggingface.co/Nekochu/Llama-2-13B-German-ORPO) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
|
58 |
+
Refer to the [original model card](https://huggingface.co/Nekochu/Llama-2-13B-German-ORPO) for more details on the model.
|
59 |
+
|
60 |
+
## Use with llama.cpp
|
61 |
+
Install llama.cpp through brew (works on Mac and Linux)
|
62 |
+
|
63 |
+
```bash
|
64 |
+
brew install llama.cpp
|
65 |
+
|
66 |
+
```
|
67 |
+
Invoke the llama.cpp server or the CLI.
|
68 |
+
|
69 |
+
### CLI:
|
70 |
+
```bash
|
71 |
+
llama-cli --hf-repo jott1970/Llama-2-13B-German-ORPO-Q4_K_M-GGUF --hf-file llama-2-13b-german-orpo-q4_k_m.gguf -p "The meaning to life and the universe is"
|
72 |
+
```
|
73 |
+
|
74 |
+
### Server:
|
75 |
+
```bash
|
76 |
+
llama-server --hf-repo jott1970/Llama-2-13B-German-ORPO-Q4_K_M-GGUF --hf-file llama-2-13b-german-orpo-q4_k_m.gguf -c 2048
|
77 |
+
```
|
78 |
+
|
79 |
+
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.
|
80 |
+
|
81 |
+
Step 1: Clone llama.cpp from GitHub.
|
82 |
+
```
|
83 |
+
git clone https://github.com/ggerganov/llama.cpp
|
84 |
+
```
|
85 |
+
|
86 |
+
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).
|
87 |
+
```
|
88 |
+
cd llama.cpp && LLAMA_CURL=1 make
|
89 |
+
```
|
90 |
+
|
91 |
+
Step 3: Run inference through the main binary.
|
92 |
+
```
|
93 |
+
./llama-cli --hf-repo jott1970/Llama-2-13B-German-ORPO-Q4_K_M-GGUF --hf-file llama-2-13b-german-orpo-q4_k_m.gguf -p "The meaning to life and the universe is"
|
94 |
+
```
|
95 |
+
or
|
96 |
+
```
|
97 |
+
./llama-server --hf-repo jott1970/Llama-2-13B-German-ORPO-Q4_K_M-GGUF --hf-file llama-2-13b-german-orpo-q4_k_m.gguf -c 2048
|
98 |
+
```
|