From the model card
Fields below are copied from the tags and counters on the HuggingFace repository meta-llama/Llama-2-7b-chat-hf at our last fetch. They are set by the uploader, not verified by us; rows with no tag are omitted. How this page is made.
- Publisher (HF namespace)
- meta-llama
- Pipeline tag
- text-generation
- Library
- Transformers
- Framework tags
- PyTorch
- Weight formats
- safetensors
- License tag
llama2— read the license file in the repo before relying on it- Language tags
- English (en)
- Papers cited
- arXiv:2307.09288
- Downloads (HF counter at last fetch)
- 508,888
- Likes (HF counter at last fetch)
- 4,823
- Model card
- https://huggingface.co/meta-llama/Llama-2-7b-chat-hf
Use cases
- Baseline for fine-tuning experiments where a reproducible 7B RLHF model is needed
- Teaching and research on instruction-tuned model behavior
- Legacy pipelines already built around LLaMA 2 weights
- Evaluating fine-tuning techniques on a well-benchmarked foundation
Pros
- Extensively benchmarked — results in literature are directly comparable
- Well-supported by every major inference framework
- LLaMA 2 license permits commercial use with restrictions
- 7B is a comfortable size for single-GPU fine-tuning
Cons
- Knowledge cutoff mid-2023; outdated facts and events
- Significantly outperformed by LLaMA 3.1 7B and later models on most benchmarks
- Context window limited to 4096 tokens
- Chat template requires explicit formatting to work correctly
Tags
transformerspytorchsafetensorsllamatext-generationfacebookmetallama-2conversationalenarxiv:2307.09288license:llama2text-generation-inferenceendpoints_compatibleregion:usdeploy:sagemaker