Update README.md
Browse files
README.md
CHANGED
@@ -15,7 +15,6 @@ tags:
|
|
15 |
### Model Description
|
16 |
|
17 |
MASHQA-Mistral-7B-Instruct is a large language model fine-tuned on healthcare question-answer pairs to respond safely to the users' queries. It is based on the Mistral-7B Instruct Architecture.
|
18 |
-
- Model creator: [Garima Sohi](https://huggingface.co/sohi-g/MASHQA-Mistral-7B-Instruct)
|
19 |
- Original model: [Mistral-7B-Instruct](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1)
|
20 |
|
21 |
### Dataset Description
|
@@ -63,17 +62,6 @@ MASHQA-Mistral-7B-Instruct is a large language model fine-tuned on healthcare qu
|
|
63 |
|
64 |
**Objective:** Safely respond to users' health-related queries.
|
65 |
|
66 |
-
### Hardware and Software - Compute Infrastructure
|
67 |
-
|
68 |
-
* 4 NVIDIA L4 GPUs & 48 vCPUs
|
69 |
-
|
70 |
-
* Environment: PyTorch (pytorch-2.0 w/ CUDA-11.8; see [Github link](https://github.com/pytorch/pytorch))
|
71 |
-
|
72 |
-
* CPU: GCP G2 Standard 48 (Platform: Intel Cascade Lake) (Accelerator Optimized)
|
73 |
-
|
74 |
-
* CPU memory: 192GB RAM
|
75 |
-
|
76 |
-
* GPU memory: 30GB per GPU
|
77 |
|
78 |
## Training
|
79 |
*This section provides information about the training.*
|
@@ -215,4 +203,4 @@ Final checkpoint after 500 epochs:
|
|
215 |
- Users of the model should provide mechanisms for those affected to provide feedback, such as an email address for comments.
|
216 |
|
217 |
# Model Card Authors
|
218 |
-
|
|
|
15 |
### Model Description
|
16 |
|
17 |
MASHQA-Mistral-7B-Instruct is a large language model fine-tuned on healthcare question-answer pairs to respond safely to the users' queries. It is based on the Mistral-7B Instruct Architecture.
|
|
|
18 |
- Original model: [Mistral-7B-Instruct](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1)
|
19 |
|
20 |
### Dataset Description
|
|
|
62 |
|
63 |
**Objective:** Safely respond to users' health-related queries.
|
64 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
65 |
|
66 |
## Training
|
67 |
*This section provides information about the training.*
|
|
|
203 |
- Users of the model should provide mechanisms for those affected to provide feedback, such as an email address for comments.
|
204 |
|
205 |
# Model Card Authors
|
206 |
+
Mohd Zeeshan
|