No special tokens for LLaMA 1 in conversion script

[ViktorooReps]

System Info

Copy-and-paste the text below in your GitHub issue and FILL OUT the two last points.

• transformers version: 4.43.3
• Platform: Linux-5.15.0-116-generic-x86_64-with-glibc2.35
• Python version: 3.12.4
• Huggingface_hub version: 0.24.3
• Safetensors version: 0.4.3
• Accelerate version: 0.33.0
• Accelerate config: - compute_environment: LOCAL_MACHINE
  - distributed_type: NO
  - mixed_precision: bf16
  - use_cpu: False
  - debug: False
  - num_processes: 1
  - machine_rank: 0
  - num_machines: 1
  - gpu_ids: all
  - rdzv_backend: static
  - same_network: True
  - main_training_function: main
  - enable_cpu_affinity: False
  - downcast_bf16: no
  - tpu_use_cluster: False
  - tpu_use_sudo: False
  - tpu_env: []
• PyTorch version (GPU?): 2.4.0+cu121 (True)
• Tensorflow version (GPU?): not installed (NA)
• Flax version (CPU?/GPU?/TPU?): not installed (NA)
• Jax version: not installed
• JaxLib version: not installed
• Using distributed or parallel set-up in script?:NO
• Using GPU in script?: NO
• GPU type: Tesla V100-SXM2-32GB

Who can help?

@ArthurZucker

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• My own task or dataset (give details below)

Reproduction

1. Download original weights to llama-tokenizer/7B
2. Run:

✗ python ../venv/lib/python3.12/site-packages/transformers/models/llama/convert_llama_weights_to_hf.py --input_dir llama-tokenizer/7B --output_dir llama7b --model_size 7B
Traceback (most recent call last):
  File "/mloscratch/homes/shcherba/landmark-attention/llama/../venv/lib/python3.12/site-packages/transformers/models/llama/convert_llama_weights_to_hf.py", line 478, in <module>
    main()
  File "/mloscratch/homes/shcherba/landmark-attention/llama/../venv/lib/python3.12/site-packages/transformers/models/llama/convert_llama_weights_to_hf.py", line 452, in main
    args.special_tokens = DEFAULT_LLAMA_SPECIAL_TOKENS[str(args.llama_version)]
                          ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^
KeyError: '1'

Expected behavior

No errors, completed conversion, as described in https://huggingface.co/docs/transformers/en/model_doc/llama

[ArthurZucker]

Hey! Indeed, we can set the default to []. As this is only used for Llama3.
Would you like to open a PR for the fix? 🤗

[ViktorooReps]

Sure, one moment