AdrianBZG commited on
Commit
e90b537
1 Parent(s): 95837d3

Upload model

Browse files
Files changed (2) hide show
  1. README.md +15 -22
  2. adapter_config.json +8 -2
README.md CHANGED
@@ -1,28 +1,21 @@
1
  ---
2
- license: apache-2.0
3
- language:
4
- - es
5
- library_name: transformers
6
- tags:
7
- - falcon
8
- - alpaca
9
- - Transformers
10
- - gpt
11
- - PyTorch
12
- - llm
13
- - llm spanish
14
- pipeline_tag: text-generation
15
- datasets:
16
- - bertin-project/alpaca-spanish
17
  ---
 
18
 
19
- <strong><span style="font-size: larger;">FALCON 7B Spanish Fine-tuned 8bit 🤗</span></strong>
20
 
21
- **Dataset**
 
 
 
 
 
 
 
 
 
 
 
22
 
23
- The dataset is a translation to Spanish of alpaca_data_cleaned.json (a clean version of the Alpaca dataset made at Stanford) using OpenAI's gpt-3.5-turbo model. This translation was made by bertin-project. It was translated using a full-sample prompt instead of per strings, which resulted in more coherent tuples of (instruction, input, output).
24
- Dataset link: [here](https://huggingface.co/datasets/bertin-project/alpaca-spanish)
25
 
26
- **Finetuning details**
27
-
28
- To fine-tune the FALCON-7B model we used the [following code](https://github.com/AdrianBZG/LLM-distributed-finetune) to run it on a distributed cluster on AWS. You are free to use such code as a fingerprint to finetune any model as you please, as it is easily customizable.
 
1
  ---
2
+ library_name: peft
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
4
+ ## Training procedure
5
 
 
6
 
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: True
10
+ - load_in_4bit: False
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
 
 
 
20
 
21
+ - PEFT 0.6.0.dev0
 
 
adapter_config.json CHANGED
@@ -1,14 +1,20 @@
1
  {
 
 
2
  "base_model_name_or_path": "tiiuae/falcon-7b",
3
  "bias": "none",
4
  "fan_in_fan_out": false,
5
  "inference_mode": true,
6
  "init_lora_weights": true,
7
- "lora_alpha": 8,
 
 
8
  "lora_dropout": 0.05,
9
  "modules_to_save": null,
10
  "peft_type": "LORA",
11
- "r": 4,
 
 
12
  "target_modules": [
13
  "query_key_value"
14
  ],
 
1
  {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
  "base_model_name_or_path": "tiiuae/falcon-7b",
5
  "bias": "none",
6
  "fan_in_fan_out": false,
7
  "inference_mode": true,
8
  "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 32,
12
  "lora_dropout": 0.05,
13
  "modules_to_save": null,
14
  "peft_type": "LORA",
15
+ "r": 16,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
  "target_modules": [
19
  "query_key_value"
20
  ],