Update README.md
Browse files
README.md
CHANGED
|
@@ -16,10 +16,12 @@ model-index:
|
|
| 16 |
This version of OpenLongCoT-Base-Gemma2-2B has been converted to run on the RK3588 NPU using ['w8a8'] quantization.
|
| 17 |
Only w8a8 quantization appears to work with Gemma 2 models. Other types throw error:
|
| 18 |
|
|
|
|
| 19 |
E RKNN: [00:14:18.994] failed to allocate handle, ret: -1, errno: 14, errstr: Bad address
|
| 20 |
E RKNN: [00:14:18.994] failed to malloc npu memory, size: 232128512, flags: 0x2
|
| 21 |
E RKNN: [00:14:18.994] load model file error!
|
| 22 |
rknn_init fail! ret=-1
|
|
|
|
| 23 |
|
| 24 |
This model has been optimized with the following LoRA:
|
| 25 |
|
|
|
|
| 16 |
This version of OpenLongCoT-Base-Gemma2-2B has been converted to run on the RK3588 NPU using ['w8a8'] quantization.
|
| 17 |
Only w8a8 quantization appears to work with Gemma 2 models. Other types throw error:
|
| 18 |
|
| 19 |
+
```
|
| 20 |
E RKNN: [00:14:18.994] failed to allocate handle, ret: -1, errno: 14, errstr: Bad address
|
| 21 |
E RKNN: [00:14:18.994] failed to malloc npu memory, size: 232128512, flags: 0x2
|
| 22 |
E RKNN: [00:14:18.994] load model file error!
|
| 23 |
rknn_init fail! ret=-1
|
| 24 |
+
```
|
| 25 |
|
| 26 |
This model has been optimized with the following LoRA:
|
| 27 |
|