You are correct that variable results could be a symptom of a failure to generalise well beyond the training set.
Such failure could happen if the models were overfit, or for other reasons. I don't think 'overfit', which is pretty well defined, is exactly the word you mean to use here.
However, I respectfully disagree with your claim. I think they are generalising well beyond the training dataset (though not as far beyond as say a good programmer would - at least not yet). I further think they are learning semantically.
Can't prove it in a comment except to say that there's simply no way they'd be able to successfully manipulate such large pieces of code, using English language instructions, it they weren't great at generalisation and ok at understanding semantics.
Such failure could happen if the models were overfit, or for other reasons. I don't think 'overfit', which is pretty well defined, is exactly the word you mean to use here.
However, I respectfully disagree with your claim. I think they are generalising well beyond the training dataset (though not as far beyond as say a good programmer would - at least not yet). I further think they are learning semantically.
Can't prove it in a comment except to say that there's simply no way they'd be able to successfully manipulate such large pieces of code, using English language instructions, it they weren't great at generalisation and ok at understanding semantics.