40 lines
917 B
Markdown
40 lines
917 B
Markdown
---
|
|
library_name: transformers
|
|
base_model: Qwen/Qwen3-0.6B
|
|
tags: [solo, fine-tuned, lora, unsloth]
|
|
datasets: [openai/gsm8k]
|
|
pipeline_tag: text-generation
|
|
---
|
|
|
|
<a href="https://hub.getsolo.tech"><img src="https://raw.githubusercontent.com/GetSoloTech/solo-cli/main/media/solo-banner.png" alt="Solo" width="200"></a>
|
|
|
|
## Model Details
|
|
|
|
| | |
|
|
|---|---|
|
|
| **Base Model** | [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B) |
|
|
| **Method** | LoRA (PEFT) |
|
|
| **Parameters** | 0.6B |
|
|
|
|
## Training Hyperparameters
|
|
|
|
| | |
|
|
|---|---|
|
|
| **Epochs** | 2 |
|
|
| **Max Steps** | 100 |
|
|
| **Batch Size** | 2 |
|
|
| **Gradient Accumulation** | 4 |
|
|
| **Learning Rate** | 0.0002 |
|
|
| **LoRA r** | 4 |
|
|
| **LoRA Alpha** | 4 |
|
|
| **Max Sequence Length** | 2048 |
|
|
| **Training Duration** | 3m |
|
|
|
|
## Dataset
|
|
|
|
[openai/gsm8k](https://huggingface.co/datasets/openai/gsm8k)
|
|
|
|
---
|
|
|
|
<sub>Trained with <a href="https://hub.getsolo.tech">Solo</a></sub>
|