From 52fe3653e59d07f88dd934a4aadad6aef4055662 Mon Sep 17 00:00:00 2001 From: David A Roberts Date: Fri, 18 Aug 2023 18:40:25 +1000 Subject: [PATCH] Fix vocab_source in sample.py --- sample.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sample.py b/sample.py index d2f56ea..c203a8c 100644 --- a/sample.py +++ b/sample.py @@ -52,7 +52,7 @@ if compile: model = torch.compile(model) # requires PyTorch 2.0 (optional) # load the tokenizer -vocab_source = checkpoint_dict.get("vocab_source", "llama2") +vocab_source = checkpoint_dict["config"].get("vocab_source", "llama2") vocab_size = gptconf.vocab_size if tokenizer: # a specific tokenizer is provided, use it