bofenghuang commited on
Commit
898b4be
1 Parent(s): 28f65c1

Initial commit

Browse files
.gitignore ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ checkpoint-*/
2
+
3
+ tmp*
README.md ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: openrail
3
+ language:
4
+ - fr
5
+ pipeline_tag: text-generation
6
+ library_name: transformers
7
+ tags:
8
+ - llama
9
+ - LLM
10
+ datasets:
11
+ - tatsu-lab/alpaca
12
+ inference: false
13
+ ---
14
+
15
+ <p align="center" width="100%">
16
+ <img src="https://huggingface.co/bofenghuang/vigogne-chat-7b/resolve/main/vigogne_logo.png" alt="Vigogne" style="width: 40%; min-width: 300px; display: block; margin: auto;">
17
+ </p>
18
+
19
+ # Vigogne-Chat-7B: A French Chat LLaMA Model
20
+
21
+ Vigogne-Chat-7B is a LLaMA-7B model fine-tuned to conduct multi-turn dialogues in 🇫🇷 French between human user and AI assistant.
22
+
23
+ For more information, please visit the Github repo: https://github.com/bofenghuang/vigogne
24
+
25
+ **Usage and License Notices**: Same as [Stanford Alpaca](https://github.com/tatsu-lab/stanford_alpaca), Vigogne is intended and licensed for research use only. The dataset is CC BY NC 4.0 (allowing only non-commercial use) and models trained using the dataset should not be used outside of research purposes.
26
+
27
+ ## Usage
28
+
29
+ This repo only contains the low-rank adapter. In order to access the complete model, you also need to load the base LLM model and tokenizer.
30
+
31
+ ```python
32
+ from peft import PeftModel
33
+ from transformers import LlamaForCausalLM, LlamaTokenizer
34
+
35
+ base_model_name_or_path = "name/or/path/to/hf/llama/7b/model"
36
+ lora_model_name_or_path = "bofenghuang/vigogne-chat-7b"
37
+
38
+ tokenizer = LlamaTokenizer.from_pretrained(base_model_name_or_path, padding_side="right", use_fast=False))
39
+ model = LlamaForCausalLM.from_pretrained(
40
+ base_model_name_or_path,
41
+ load_in_8bit=True,
42
+ torch_dtype=torch.float16,
43
+ device_map="auto",
44
+ )
45
+ model = PeftModel.from_pretrained(model, lora_model_name_or_path)
46
+ ```
47
+
48
+ <!-- You can infer this model by using the following Google Colab Notebook.
49
+
50
+ <a href="https://colab.research.google.com/github/bofenghuang/vigogne/blob/main/notebooks/infer_instruct.ipynb" target="_blank"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/></a> -->
51
+
52
+ ## Limitations
53
+
54
+ Vigogne is still under development, and there are many limitations that have to be addressed. Please note that it is possible that the model generates harmful or biased content, incorrect information or generally unhelpful answers.
adapter_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "base_model_name_or_path": "huggyllama/llama-7b",
3
+ "bias": "none",
4
+ "fan_in_fan_out": false,
5
+ "inference_mode": true,
6
+ "init_lora_weights": true,
7
+ "lora_alpha": 16,
8
+ "lora_dropout": 0.05,
9
+ "modules_to_save": null,
10
+ "peft_type": "LORA",
11
+ "r": 8,
12
+ "target_modules": [
13
+ "q_proj",
14
+ "v_proj",
15
+ "k_proj",
16
+ "o_proj",
17
+ "gate_proj",
18
+ "down_proj",
19
+ "up_proj"
20
+ ],
21
+ "task_type": "CAUSAL_LM"
22
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17a8c8bf51e0cd3fbacd1391c60445844ac65f20b6ac3617d99ea13de542a737
3
+ size 40047549
runs/1/1683884762.493527/events.out.tfevents.1683884762.koios.zaion.ai.224530.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37130027386eb12a9a683dcf5e52df1bbd2d948e93eafa9f2c3381e339134034
3
+ size 6010
runs/1/1683888216.0705674/events.out.tfevents.1683888216.koios.zaion.ai.234743.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5daf3014f1d749ef42e5bd7d69ffed5279f81b7d77b2e89ba9615f679990826
3
+ size 6010
runs/1/1683893771.5868623/events.out.tfevents.1683893771.koios.zaion.ai.249409.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ede703f88b2c0280493972a833791e0a5eae1742b8abfd08c8cd8f55d8e4c05b
3
+ size 6010
runs/1/events.out.tfevents.1683884762.koios.zaion.ai.224530.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5874cd25670604e5413619d1be4be6f21a45dd191d8fabbd830dd67ae236231c
3
+ size 4056
runs/1/events.out.tfevents.1683888216.koios.zaion.ai.234743.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:65c312df17a5edf956db8dd8511b48e564d65335f42cac0f0be667565c317637
3
+ size 5442
runs/1/events.out.tfevents.1683893771.koios.zaion.ai.249409.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce50e905326e06cf4cbbdcd0039fe63b6888ac38f523b2a839baa139b19f10de
3
+ size 53017
vigogne_logo.png ADDED