divyanshukunwar commited on
Commit
36aca1b
·
verified ·
1 Parent(s): 50681c5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +28 -7
README.md CHANGED
@@ -1,19 +1,18 @@
1
  ---
2
  license: apache-2.0
3
  language:
 
4
  - hi
5
  - en
6
- - bh
7
  base_model:
8
  - google/gemma-2-9b
9
  library_name: transformers
10
- tags:
11
- - unsloth
12
  ---
 
 
 
 
13
 
14
- SASTRI_1_9B is a new model fine tuned for hindi & bhojpuri languages.
15
- More detail about the dataset and evaluation result will be shared soon.
16
- Model trained on similar languages can be adapted for a language that is not officially supported.
17
  Let's see what model has to say about itself.
18
 
19
  You can also download the original safetensor version from : https://huggingface.co/divyanshukunwar/SASTRI_1_9B
@@ -24,4 +23,26 @@ As a large language model, I don't have personal feelings or emotions like human
24
 
25
  User : aap or kon kon si bhaasa me likh skte hai ?
26
  Assistant : मैं कई भाषाओं को समझता हूँ और उनमें बातचीत कर सकता हूँ, हिंदी में लिखने के अलावा मैं अंग्रेजी और भोजपुरी भी जानता हूं। 🇮🇳
27
- ```
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
  language:
4
+ - bh
5
  - hi
6
  - en
 
7
  base_model:
8
  - google/gemma-2-9b
9
  library_name: transformers
 
 
10
  ---
11
+ This is GGUF version of the model.
12
+ SASTRI_1_9B is a new model fine tuned for bhojpuri & hindi languages.
13
+ More detail about the dataset used and evaluation will be shared soon.
14
+ Model trained on similar languages can be adapted for a very low resource language.
15
 
 
 
 
16
  Let's see what model has to say about itself.
17
 
18
  You can also download the original safetensor version from : https://huggingface.co/divyanshukunwar/SASTRI_1_9B
 
23
 
24
  User : aap or kon kon si bhaasa me likh skte hai ?
25
  Assistant : मैं कई भाषाओं को समझता हूँ और उनमें बातचीत कर सकता हूँ, हिंदी में लिखने के अलावा मैं अंग्रेजी और भोजपुरी भी जानता हूं। 🇮🇳
26
+ ```
27
+
28
+ Evaluation Result for English Language
29
+ |Dataset| Metric | Score |
30
+ |--------|---------|---------|
31
+ |Arc-Easy| Accuracy | 94.2 |
32
+ |Arc-Challenge| Accuracy | 86.1 |
33
+ |MMLU| Accuracy | 61.0 |
34
+ |BoolQ| Accuracy | 79.04 |
35
+
36
+ Evaluation Result for Hindi Language
37
+ |Dataset| Metric | Score |
38
+ |--------|---------|---------|
39
+ |Arc-Easy| Accuracy | 80.2 |
40
+ |Arc-Challenge| Accuracy | 68.35 |
41
+ |MMLU| Accuracy | 43.11 |
42
+ |BoolQ| Accuracy | 71.61 |
43
+
44
+ Evaluation Result for Bhojpuri Language coming soon ..
45
+
46
+ Model may so undersired behavior such as hallucinations and unintended biases use it at your own risk.
47
+ This model is not optimized for user preference.
48
+ This model is released as a part of study for Large Language model for very low resource languages.