|
--- |
|
library_name: peft |
|
license: apache-2.0 |
|
datasets: |
|
- lucasmccabe-lmi/CodeAlpaca-20k |
|
language: |
|
- en |
|
tags: |
|
- code |
|
--- |
|
# Model Card for DeciCoder 1B - CodeAlpaca20k Fine Tuned |
|
|
|
This model is a fine tuned version of DeciCoder 1B (https://huggingface.co/Deci/DeciCoder-1b), fine tune for instructions based on Code Alpaca Dataset. |
|
|
|
## 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: True |
|
- bnb_4bit_compute_dtype: float16 |
|
|
|
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: True |
|
- bnb_4bit_compute_dtype: float16 |
|
### Framework versions |
|
|
|
- PEFT 0.5.0 |
|
|
|
## Citation |
|
Thanks for DeciCoder-1b team for making this model open sourced. |
|
``` |
|
@misc{DeciFoundationModels, |
|
title = {DeciCoder}, |
|
author = {DeciAI Research Team}, |
|
year = {2023} |
|
url={[https://huggingface.co/deci/decicoder-1b](https://huggingface.co/deci/decicoder-1b)}, |
|
} |
|
``` |
