Instructions to use ibm-ai-platform/llama3-70b-accelerator with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ibm-ai-platform/llama3-70b-accelerator with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("ibm-ai-platform/llama3-70b-accelerator", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download config.json from ibm-ai-platform/llama3-70b-accelerator: direct link, hf CLI and curl.
- Browser
- Download file 441 Bytes
-
https://huggingface.co/ibm-ai-platform/llama3-70b-accelerator/resolve/main/config.json
- Command line
-
hf download hf://ibm-ai-platform/llama3-70b-accelerator/config.json
-
curl -L -o config.json https://huggingface.co/ibm-ai-platform/llama3-70b-accelerator/resolve/main/config.json
441 Bytes
| { | |
| "base_model_name_or_path": "meta-llama/Meta-Llama-3-70B-Instruct", | |
| "architectures": [ | |
| "MLPSpeculatorPreTrainedModel" | |
| ], | |
| "emb_dim": 8192, | |
| "inner_dim": 8192, | |
| "model_type": "mlp_speculator", | |
| "n_candidates": 4, | |
| "n_predict": 4, | |
| "scale_input": true, | |
| "tie_weights": true, | |
| "top_k_tokens_per_head": [ | |
| 4, | |
| 3, | |
| 2, | |
| 2 | |
| ], | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.41.2", | |
| "vocab_size": 128256 | |
| } | |