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 by adding defaulting to encoder + Fix for sequence length reading from MTEB

fixed NaNs

Update modeling_gptbert.py

make FlashAttention logic more robust

Upload model.safetensors with huggingface_hub

fix

Update config.json

removed SDPA

Update modeling_gptbert.py

fixed SDPA for older PyTorch versions

Delete model.safetensors

Adding `safetensors` variant of this model (#12)

Upload model_performance.png

Update README.md

Update config.json

fix

Update modeling_gptbert.py

Update modeling_gptbert.py

fix

fix

fix

fix attention mask

fix

fix

Update modeling_gptbert.py

Update configuration_gptbert.py

Update modeling_gptbert.py

Update config.json

Update config.json

fixes and optimizations

Fix import error

Update modeling_gptbert.py

Update modeling_gptbert.py

Update modeling_gptbert.py

Update modeling_gptbert.py

Update modeling_gptbert.py

Update modeling_gptbert.py

Update modeling_gptbert.py

Update modeling_gptbert.py

Checking for errors in the CastedLinear

Update modeling_gptbert.py

More testing

Add checking for NaN

Removing added print

Testing output

fix flasshattention