intervitens
intervitens
/ tokenizer_fix.py
Last active
January 26, 2024 22:20
Script for the InternLM2 tokenizer to add ChatML tokens and fix null token 354 for ggml conversion
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Launch with PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python python tokenizer_fix.py | |
| import sentencepiece.sentencepiece_model_pb2 as model | |
| m = model.ModelProto() | |
| m.ParseFromString(open('./tokenizer.model', 'rb').read()) | |
| m.pieces[92543].piece = '<|im_start|>' | |
| m.pieces[92542].piece = '<|im_end|>' | |
| m.pieces[92541].piece = '<|action_start|>' | |
| m.pieces[92540].piece = '<|action_end|>' |