h2oGPT Model Card Summary H2O.ai's h2ogpt oig oasst1 512 6 9b is a 6.9 billion parameter instruction following large language model licensed for commercial use. Base model: EleutherAI/pythia 6.9b Fine tuning dataset: h2oai/h2ogpt oig oasst1 instruct cleaned v1 and h2oai/openassistant oasst1 h2ogpt and h2oai/h2ogpt fortune2000 personalized and h2oai/h2ogpt oig oasst1 instruct cleaned v3 Data prep and fine tuning code: H2O.ai GitHub Training logs: zip and zip and zip Chatbot Run your own chatbot: H2O.ai GitHub Usage To use the model with the transformers library on a machine with GPUs, first make sure you have the transformers and accelerate libraries installed. Alternatively, if you prefer to not use trust remote code=True you can download instruct pipeline.py, store it alongside your notebook, and construct the pipeline yourself from the loaded model and tokenizer: Model Architecture Model Configuration Model Validation Model validation results using EleutherAI lm evaluation harness. eval source code Task Version Metric Value Stderr : : : arc easy 0 acc 0.6591 ± 0.0097 acc norm 0.6178 ± 0.0100 arc challenge 0 acc 0.3174 ± 0.0136 acc norm 0.3558 ± 0.0140 openbookqa 0 acc 0.2540 ± 0.…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy