Update doc for grad_accu and add validation tests for batch size 3c71c8d Nanobit commited on May 31, 2023
fix packing so that concatenated sequences reset the attention 9b8585d winglian commited on May 31, 2023
new hf_use_auth_token setting so login to hf isn't required 1c33eb8 winglian commited on May 28, 2023