This commit is contained in:
Chime Ogbuji 2024-11-06 20:29:12 -05:00 committed by Awni Hannun
parent 02abeeade4
commit 71d9f8cc38

View File

@ -166,7 +166,7 @@ def iterate_delineated_batches(
for j in batch_idx[i]:
prompt, completion = dataset.get_prompt_and_completion(j)
prompt_lengths.append(input_length(prompt, completion, tokenizer))
full_sequence = [tokenizer.encode(dataset[j]) for j in batch_idx[i]]
full_sequence = tokenizer.encode(dataset[j])
if full_sequence[-1] != tokenizer.eos_token_id:
full_sequence.append(tokenizer.eos_token_id)
batch.append(full_sequence)