lxs1
/

DistilBertForSequenceClassification_6h_768dim

@@ -14,7 +14,7 @@
 - **Known limitations**: The model may exhibit biases present in the training data, potentially leading to inaccuracies in certain contexts or for specific demographic groups. Its performance has not been extensively tested across all possible domains, so results may vary for texts outside of the training distribution.
 ## Hardware
-- **Training Platform**: The model was trained on a cloud computing platform using NVIDIA Tesla V100 GPUs. Training involved multiple epochs over the dataset with careful monitoring for overfitting.
 ## Software Optimizations
 - **Known Optimizations**: During training, techniques such as gradient accumulation and mixed-precision training were employed to enhance performance and reduce memory usage. The AdamW optimizer was used for its effective learning rate adjustments.

 - **Known limitations**: The model may exhibit biases present in the training data, potentially leading to inaccuracies in certain contexts or for specific demographic groups. Its performance has not been extensively tested across all possible domains, so results may vary for texts outside of the training distribution.
 ## Hardware
+- **Training Platform**: The model was trained on Intel Developer Cloud over scalable  Intel® Xeon® 4th Gen Scalable processors.
 ## Software Optimizations
 - **Known Optimizations**: During training, techniques such as gradient accumulation and mixed-precision training were employed to enhance performance and reduce memory usage. The AdamW optimizer was used for its effective learning rate adjustments.