Modell Meta-Llama-3-70B

Models
1

I would like to download the “Code Llama-3-70B-Instruct” model to improve my Python skills. Is there a way to integrate this into my Debian system, or to download it? Thanks for helping

2

Hi,

There are various ways to use the model.

  • easiest is to using it with the Transformers library as shown in the model card: meta-llama/CodeLlama-70b-Instruct-hf · Hugging Face (see the Python code snippets). Make sure to accept the form presented there as Meta requires you to share your contact information before using the model. If you use the model with Transformers (which you can install with pip install transformers), it’s going to load the weights in a local cache directory.
  • if you just want to load the weights locally you can git clone the files of the repository. Click the 3 dots on the right and select “clone this repository”.
3

Hello again, thank you for the help. The model was downloaded up to 29/29, then it stopped working. While I was looking for the problem, I saw the 92b model and thought I would try it, but I encountered many error messages and am at a loss. It is important for me to learn a lot about the Python programming language… I thank you very much for your valuable time. Here are the error messages:

/home/user/Llama/Llama-3-70B/venv/bin/python /home/user/PyrePython/Pure Python/download_model.py 
Downloading model nisten/BigCodeLlama-92b...
loading file tokenizer.model from cache at /home/user/.cache/huggingface/hub/models--nisten--BigCodeLlama-92b/snapshots/5a3ecf5b143fae6373ac18c5a7cbcc3d12ece200/tokenizer.model
loading file tokenizer.json from cache at /home/user/.cache/huggingface/hub/models--nisten--BigCodeLlama-92b/snapshots/5a3ecf5b143fae6373ac18c5a7cbcc3d12ece200/tokenizer.json
loading file added_tokens.json from cache at None
loading file special_tokens_map.json from cache at /home/user/.cache/huggingface/hub/models--nisten--BigCodeLlama-92b/snapshots/5a3ecf5b143fae6373ac18c5a7cbcc3d12ece200/special_tokens_map.json
loading file tokenizer_config.json from cache at /home/user/.cache/huggingface/hub/models--nisten--BigCodeLlama-92b/snapshots/5a3ecf5b143fae6373ac18c5a7cbcc3d12ece200/tokenizer_config.json
loading configuration file config.json from cache at /home/user/.cache/huggingface/hub/models--nisten--BigCodeLlama-92b/snapshots/5a3ecf5b143fae6373ac18c5a7cbcc3d12ece200/config.json
Model config LlamaConfig {
  "_name_or_path": "nisten/BigCodeLlama-92b",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 8192,
  "initializer_range": 0.02,
  "intermediate_size": 28672,
  "max_position_embeddings": 2048,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 64,
  "num_hidden_layers": 107,
  "num_key_value_heads": 8,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.41.2",
  "use_cache": true,
  "vocab_size": 32016
}

loading weights file model.safetensors from cache at /home/user/.cache/huggingface/hub/models--nisten--BigCodeLlama-92b/snapshots/5a3ecf5b143fae6373ac18c5a7cbcc3d12ece200/model.safetensors.index.json
Downloading shards:   0%|          | 0/19 [00:00<?, ?it/s]