| library_name: peft | |
| base_model: mistralai/Mistral-7B-v0.1 | |
| pipeline_tag: text-generation | |
| Description: Multiple-choice sentence completion\ | |
| Original dataset: https://huggingface.co/datasets/Rowan/hellaswag \ | |
| ---\ | |
| Try querying this adapter for free in Lora Land at https://predibase.com/lora-land! \ | |
| The adapter_category is Other and the name is Multiple Choice Sentence Completion (hellaswag)\ | |
| ---\ | |
| Sample input: You are provided with an incomplete passage below as well as 4 endings in quotes and separated by commas, with only one of them being the correct ending. Treat the endings as being labelled 0, 1, 2, 3 in order. Please respond with the number corresponding to the correct ending for the passage.\n\n### Passage: The mother instructs them on how to brush their teeth while laughing. The boy helps his younger sister brush his teeth. she\n\n### Endings: ['shows how to hit the mom and then kiss his dad as well.' | |
| 'brushes past the camera, looking better soon after.' | |
| 'glows from the center of the camera as a reaction.' | |
| 'gets them some water to gargle in their mouths.']\n\n### Correct Ending Number: \ | |
| ---\ | |
| Sample output: 3.0\ | |
| ---\ | |
| Try using this adapter yourself! | |
| ``` | |
| from transformers import AutoModelForCausalLM, AutoTokenizer | |
| model_id = "mistralai/Mistral-7B-v0.1" | |
| peft_model_id = "predibase/hellaswag" | |
| model = AutoModelForCausalLM.from_pretrained(model_id) | |
| model.load_adapter(peft_model_id) | |
| ``` |