chainyo
commited on
Commit
•
38a13e0
1
Parent(s):
31458d1
add training hardware
Browse files
README.md
CHANGED
|
@@ -176,3 +176,7 @@ The performance degradation is due to the fact we load the model in 8bit and we
|
|
| 176 |
Thanks to the 8bit quantization, the model is 4 times faster than the original model and the results are still decent.
|
| 177 |
|
| 178 |
Some complex tasks like WinoGrande and OpenBookQA are more difficult to solve with the adapters.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 176 |
Thanks to the 8bit quantization, the model is 4 times faster than the original model and the results are still decent.
|
| 177 |
|
| 178 |
Some complex tasks like WinoGrande and OpenBookQA are more difficult to solve with the adapters.
|
| 179 |
+
|
| 180 |
+
## Training Hardware
|
| 181 |
+
|
| 182 |
+
This model was trained on a GCP instance with 16x NVIDIA A100 40GB GPUs.
|