Hugging Face's logo

MohammedNasser
/
silma_9b_instruct_ft

Question Answering
Transformers
Safetensors
Arabic
Model card Files
xet
Community
MohammedNasser commited on
Commit
92e1520
·
verified ·
1 Parent(s): 346528f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -6
README.md CHANGED
@@ -13,18 +13,19 @@ pipeline_tag: question-answering
13
 
14
  # SILMA-9B-Instruct Fine-Tuned for Arabic QA
15
 
16
- ![Model Banner](https://your-banner-image-url.jpg)
17
 
18
- [![Generic badge](https://img.shields.io/badge/🤗-Hugging%20Face-blue.svg)](https://huggingface.co/MohammedNasser/silma_9b_instruct_ft)
19
- [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
20
- [![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/release/python-390/)
21
 
22
  This model is a fine-tuned version of [silma-ai/SILMA-9B-Instruct-v1.0](https://huggingface.co/silma-ai/SILMA-9B-Instruct-v1.0), optimized for Arabic Question Answering tasks. It excels at providing numerical answers to a wide range of questions in Arabic.
23
 
24
  ## Model Details
25
 
 
 
 
 
 
26
  - **Task**: Arabic Question Answering (Numerical Responses)
27
- - **Fine-tuning Technique**: LoRA (Low-Rank Adaptation)
28
  - **Training Data**: [Custom Arabic Reasoning QA dataset](https://huggingface.co/MohammedNasser/ARabic_Reasoning_QA)
29
  - **Quantization**: 4-bit quantization using bitsandbytes
30
 
@@ -33,7 +34,7 @@ This model is a fine-tuned version of [silma-ai/SILMA-9B-Instruct-v1.0](https://
33
  - Optimized for Arabic language understanding and generation
34
  - Specialized in providing numerical answers to questions
35
  - Efficient inference with 4-bit quantization
36
- - Fine-tuned using LoRA for parameter-efficient training
37
 
38
  Step |Training Loss |Validation Loss
39
  10 2.207200 1.487218
 
13
 
14
  # SILMA-9B-Instruct Fine-Tuned for Arabic QA
15
 
 
16
 
17
+ [![Generic badge](https://img.shields.io/badge/🤗-Hugging%20Face-blue.svg)](https://huggingface.co/MohammedNasser/silma_9b_instruct_ft)[![License: Apache2.0](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)[![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/release/python-390/)
 
 
18
 
19
  This model is a fine-tuned version of [silma-ai/SILMA-9B-Instruct-v1.0](https://huggingface.co/silma-ai/SILMA-9B-Instruct-v1.0), optimized for Arabic Question Answering tasks. It excels at providing numerical answers to a wide range of questions in Arabic.
20
 
21
  ## Model Details
22
 
23
+ - **Model Name**: silma_9b_instruct_ft
24
+ - **Model Type**: Language Model
25
+ - **Language**: Arabic
26
+ - **Base Model**: silma-ai/SILMA-9B-Instruct-v1.0
27
+ - **Fine-Tuning Method**: PEFT with LoraConfig
28
  - **Task**: Arabic Question Answering (Numerical Responses)
 
29
  - **Training Data**: [Custom Arabic Reasoning QA dataset](https://huggingface.co/MohammedNasser/ARabic_Reasoning_QA)
30
  - **Quantization**: 4-bit quantization using bitsandbytes
31
 
 
34
  - Optimized for Arabic language understanding and generation
35
  - Specialized in providing numerical answers to questions
36
  - Efficient inference with 4-bit quantization
37
+ - Fine-tuned using PEFT with LoraConfig for parameter-efficient training
38
 
39
  Step |Training Loss |Validation Loss
40
  10 2.207200 1.487218