komus/physicase_tuned_gemma3

---

base_model: google/gemma-3-1b-pt language: en license: apache-2.0

Fine-tuned Model with LoRA

Model Description

This model is a fine-tuned.

Training Configuration

LoRA Parameters

• Rank (r): 4
• Alpha: 8
• Dropout: 0.05
• Target Modules: ['q_proj', 'v_proj']

Training Hyperparameters

• Learning Rate: 0.0001
• Epochs: 3
• Batch Size: 1
• Gradient Accumulation Steps: 4
• Optimizer: AdamW
• Weight Decay: 0.01

Training Data

• Dataset: Physicase
• Training Samples: 318

Performance

[Include metrics here]