ace2105/llama2-api-test

Model Card for Model ID

Llama2 model fine tuned on different API's for generating test cases.

Model Details

Model Description

As a developer, testing is a major concern. Keeping this in mind, came up with this fine tuned LLaMA-2-7b on different API. Through this model you can upload your API's detail and test cases scenarios will be generated.

• Developed by: Anish Vantagodi
• Funded by : Kusho
• Shared by : Anish Vantagodi
• Model type: LlaMA-v7 Peft fine tuned
• Language(s) (NLP): English
• License: MIT
• Finetuned from model [optional]: meta-llama/Llama-2-7b-hf

Uses

Used for generating and testing API's

Model Card Authors

Anish Vantagodi: https://github.com/anish2105

Training procedure

The following bitsandbytes quantization config was used during training:

• quant_method: bitsandbytes
• load_in_8bit: False
• load_in_4bit: True
• llm_int8_threshold: 6.0
• llm_int8_skip_modules: None
• llm_int8_enable_fp32_cpu_offload: False
• llm_int8_has_fp16_weight: False
• bnb_4bit_quant_type: nf4
• bnb_4bit_use_double_quant: False
• bnb_4bit_compute_dtype: float16

Framework versions

• PEFT 0.6.2