ss-76
/

microgpt-deva

Text Generation

Model card Files Files and versions

ss-76 commited on Jul 30

Commit

b2c8e08

·

verified ·

1 Parent(s): 796287d

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -45,7 +45,7 @@ This model is ideal for:
 - **Data**: Custom Sanskrit dataset of over 100,000+ Devanagari `.txt` files.
 - **Tokenizer**: [SentencePiece](https://github.com/google/sentencepiece) BPE model trained with `character_coverage=1.0`.
-- **Training Platform**: AWS SageMaker (`ml.p3.2xlarge`)
 - **Framework**: PyTorch with custom FlashAttention blocks
 - **Training Time**: ~3 epochs with dynamic batching on sharded data

 - **Data**: Custom Sanskrit dataset of over 100,000+ Devanagari `.txt` files.
 - **Tokenizer**: [SentencePiece](https://github.com/google/sentencepiece) BPE model trained with `character_coverage=1.0`.
+- **Training Platform**: AWS SageMaker Tesla V100 GPU
 - **Framework**: PyTorch with custom FlashAttention blocks
 - **Training Time**: ~3 epochs with dynamic batching on sharded data