Model Card

Model Details

  • Base Model: Qwen/Qwen2.5-7B
  • Dataset: cloudwalk-kickass/inkman_label_extention_gpt4o_mini
  • Dataset size: 18000
  • Revision: 96d1f08

Training

The label used is : This model was fine-tuned using the following hyperparameters:

  • Learning rate: 1e-05
  • Batch size: 4
  • Max epochs: 2
  • Weight decay: 0.01

Prompt

  • Prefix: You are tasked with analyzing structured credit reports and predicting whether a merchant is likely to repay a future loan. Each report contains detailed features about the merchant's financial performance, platform engagement, risk indicators, and customer interactions. Your goal is to evaluate the provided information and output a binary prediction: 'True' if the merchant is likely to repay the loan, or 'False' if the merchant is unlikely to repay.

  • Suffix: Based on the above information, describe the merchant's financial situation in 3 short sentences that could help you predict this binary repayment outcome. Indicate the binary repayment outcome after the sentences. The sentences and the binary repayment outcome should be separated by a new line.

Downloads last month
50
Safetensors
Model size
7.62B params
Tensor type
BF16
·
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and HF Inference API was unable to determine this model's library.