29 lines
621 B
Markdown
29 lines
621 B
Markdown
|
---
|
||
|
|
base_model: unsloth/qwen3-4b-base-unsloth-bnb-4bit
|
||
|
|
tags:
|
||
|
|
- text-generation-inference
|
||
|
|
- transformers
|
||
|
|
- unsloth
|
||
|
|
- qwen3
|
||
|
|
license: apache-2.0
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
datasets:
|
||
|
|
- ajibawa-2023/Go-Code-Large
|
||
|
|
---
|
||
|
|
|
||
|
|
# Qwen3-Go
|
||
|
|
|
||
|
|
An attempt at local finetuning using a 5070 TI.
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
Qwen3-Go is a model finetuned using the [G0-Code-Large](https://huggingface.co/datasets/ajibawa-2023/Go-Code-Large) dataset for Go code completion.
|
||
|
|
|
||
|
|
This is a purely experimental model.
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
### Training Parameters
|
||
|
|
|
||
|
|
This model is trained locally on a 5070Ti using Unsloth loading in 4bits. With a batch size of 8 and 1 epoch. The training took 19 hours.
|