Update README.md
Browse files
README.md
CHANGED
|
@@ -49,12 +49,28 @@ language:
|
|
| 49 |
metrics:
|
| 50 |
- accuracy
|
| 51 |
base_model:
|
| 52 |
-
-
|
| 53 |
pipeline_tag: question-answering
|
| 54 |
tags:
|
| 55 |
- biology
|
| 56 |
- medical
|
| 57 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
# Democratizing Medical LLMs For Much More Languages
|
| 59 |
|
| 60 |
Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish, Arabic, Russian, Japanese, Korean, German, Italian, Portuguese and 38 Minor Languages So far.
|
|
@@ -67,7 +83,7 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
|
|
| 67 |
|
| 68 |
|
| 69 |
|
| 70 |
-

|
| 71 |
|
| 72 |
|
| 73 |
## 🌈 Update
|
|
@@ -81,7 +97,7 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
|
|
| 81 |
<details>
|
| 82 |
<summary>Click to view the Languages Coverage</summary>
|
| 83 |
|
| 84 |
-

|
| 85 |
|
| 86 |
</details>
|
| 87 |
|
|
@@ -91,7 +107,7 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
|
|
| 91 |
<details>
|
| 92 |
<summary>Click to view the MoE routing image</summary>
|
| 93 |
|
| 94 |
-

|
| 95 |
|
| 96 |
</details>
|
| 97 |
|
|
@@ -105,7 +121,7 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
|
|
| 105 |
<details>
|
| 106 |
<summary>Click to view the Dense Models Results</summary>
|
| 107 |
|
| 108 |
-

|
| 109 |
|
| 110 |
</details>
|
| 111 |
|
|
@@ -116,7 +132,7 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
|
|
| 116 |
<details>
|
| 117 |
<summary>Click to view the Post-MoE Models Results</summary>
|
| 118 |
|
| 119 |
-

|
| 120 |
|
| 121 |
</details>
|
| 122 |
|
|
@@ -139,7 +155,7 @@ Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish,
|
|
| 139 |
|
| 140 |
<details><summary>Click to expand</summary>
|
| 141 |
|
| 142 |
-

|
| 143 |
|
| 144 |
- [Data category](https://huggingface.co/datasets/FreedomIntelligence/ApolloCorpus/tree/main/train)
|
| 145 |
|
|
|
|
| 49 |
metrics:
|
| 50 |
- accuracy
|
| 51 |
base_model:
|
| 52 |
+
- FreedomIntelligence/Apollo2-7B
|
| 53 |
pipeline_tag: question-answering
|
| 54 |
tags:
|
| 55 |
- biology
|
| 56 |
- medical
|
| 57 |
---
|
| 58 |
+
# Apollo2-7B-exl2
|
| 59 |
+
Original model: [Apollo2-7B](https://huggingface.co/FreedomIntelligence/Apollo2-7B)
|
| 60 |
+
Made by: [FreedomIntelligence](https://huggingface.co/FreedomIntelligence)
|
| 61 |
+
|
| 62 |
+
## Quants
|
| 63 |
+
[4bpw h6 (main)](https://huggingface.co/cgus/Apollo2-7B-exl2/tree/main)
|
| 64 |
+
[4.5bpw h6](https://huggingface.co/cgus/Apollo2-7B-exl2/tree/4.5bpw-h6)
|
| 65 |
+
[5bpw h6](https://huggingface.co/cgus/Apollo2-7B-exl2/tree/5bpw-h6)
|
| 66 |
+
[6bpw h6](https://huggingface.co/cgus/Apollo2-7B-exl2/tree/6bpw-h6)
|
| 67 |
+
[8bpw h8](https://huggingface.co/cgus/Apollo2-7B-exl2/tree/8bpw-h8)
|
| 68 |
+
|
| 69 |
+
## Quantization notes
|
| 70 |
+
Made with Exllamav2 0.2.3 with the default dataset. This model needs software with Exllamav2 library such as Text-Generation-WebUI, TabbyAPI, etc.
|
| 71 |
+
This model has to fit your GPU to be usable and it's mainly meant for RTX cards on Windows/Linux or AMD on Linux.
|
| 72 |
+
|
| 73 |
+
# Original model card
|
| 74 |
# Democratizing Medical LLMs For Much More Languages
|
| 75 |
|
| 76 |
Covering 12 Major Languages including English, Chinese, French, Hindi, Spanish, Arabic, Russian, Japanese, Korean, German, Italian, Portuguese and 38 Minor Languages So far.
|
|
|
|
| 83 |
|
| 84 |
|
| 85 |
|
| 86 |
+

|
| 87 |
|
| 88 |
|
| 89 |
## 🌈 Update
|
|
|
|
| 97 |
<details>
|
| 98 |
<summary>Click to view the Languages Coverage</summary>
|
| 99 |
|
| 100 |
+

|
| 101 |
|
| 102 |
</details>
|
| 103 |
|
|
|
|
| 107 |
<details>
|
| 108 |
<summary>Click to view the MoE routing image</summary>
|
| 109 |
|
| 110 |
+

|
| 111 |
|
| 112 |
</details>
|
| 113 |
|
|
|
|
| 121 |
<details>
|
| 122 |
<summary>Click to view the Dense Models Results</summary>
|
| 123 |
|
| 124 |
+

|
| 125 |
|
| 126 |
</details>
|
| 127 |
|
|
|
|
| 132 |
<details>
|
| 133 |
<summary>Click to view the Post-MoE Models Results</summary>
|
| 134 |
|
| 135 |
+

|
| 136 |
|
| 137 |
</details>
|
| 138 |
|
|
|
|
| 155 |
|
| 156 |
<details><summary>Click to expand</summary>
|
| 157 |
|
| 158 |
+

|
| 159 |
|
| 160 |
- [Data category](https://huggingface.co/datasets/FreedomIntelligence/ApolloCorpus/tree/main/train)
|
| 161 |
|