Using max_position_embeddings instead of max_sequence_length to standardise with HF

Using max_position_embeddings instead of max_sequence_length to standardise with HF

Fix causal mode

Update config.json

Fixing AutoModel initialization to intialize as encoder + Fixing sequence length reading for MTEB

fixed NaNs

Create generation_config.json

Update modeling_gptbert.py

Update modeling_gptbert.py

Make the FlashAttention logic more robust

fix

Update config.json

removed SDPA

Update modeling_gptbert.py

fixed SDPA for older PyTorch versions

Update README.md

Update README.md

Update README.md

Update config.json

Upload model_performance.png

Update README.md

Update config.json

FlashAttention support

FlashAttention support

FlashAttention support

Update modeling_gptbert.py

Fixes and optimizations

Update special_tokens_map.json

Update modeling_gptbert.py

Update configuration_gptbert.py

Upload folder using huggingface_hub

initial commit