|
--- |
|
library_name: peft |
|
base_model: mistralai/Mistral-7B-v0.1 |
|
pipeline_tag: text-generation |
|
--- |
|
Description: Topic identification from a news article, with synthetic explanations.\ |
|
Original dataset: https://huggingface.co/datasets/ag_news \ |
|
---\ |
|
Try querying this adapter for free in Lora Land at https://predibase.com/lora-land! \ |
|
The adapter_category is Explanation and the name is News Topic Identification Explained (ag_news)\ |
|
---\ |
|
Sample input: Below is a news article. Please classify it under one of the following classes (World, Business, Sports, Sci/Tech) and provide a reasonable coherent explanation for why the article is classified as such. Please format your response as a JSON payload.\n\n### Article: US poverty rate climbs, along with number lacking health coverage (AFP) AFP - The number of Americans living in poverty or without health insurance grew last year, a government survey showed, adding potential dynamite in the battle for the White House.\n\n### JSON Response\ |
|
---\ |
|
Sample output: {"text_label": "World", "explanation": "The article is classified as 'World' because it discusses the poverty rate and lack of health coverage in the United States. Although it is a domestic issue, it has potential implications for the upcoming presidential election, making it relevant to a global audience."}\ |
|
---\ |
|
Try using this adapter yourself! |
|
``` |
|
from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
|
model_id = "mistralai/Mistral-7B-v0.1" |
|
peft_model_id = "predibase/agnews_explained" |
|
|
|
model = AutoModelForCausalLM.from_pretrained(model_id) |
|
model.load_adapter(peft_model_id) |
|
``` |
