---
library_name: transformers
tags:
- unsloth
datasets:
- amityco/tau-bench-retail-train-next-action-medium
---
```

tau train medium 100 sample

│ 🏆 Average Reward: 0.5175                                                                                                                    │
│                                                                                                                                              │
│ 📈 Pass^k Metrics:                                                                                                                           │
│ k=1: 0.518                                                                                                                                   │
│ k=2: 0.421

tau train medium 200 sample

│ 🏆 Average Reward: 0.5526                                                                                                                    │
│                                                                                                                                              │
│ 📈 Pass^k Metrics:                                                                                                                           │
│ k=1: 0.553                                                                                                                                   │
│ k=2: 0.482   

tau train medium 300 sample

│ 🏆 Average Reward: 0.5175                                                                                                                    │
│                                                                                                                                              │
│ 📈 Pass^k Metrics:                                                                                                                           │
│ k=1: 0.518                                                                                                                                   │
│ k=2: 0.421  
```

![image](https://cdn-uploads.huggingface.co/production/uploads/64739bc371f07ae738d2d61d/-efahsrelKuAcCpzzGrtv.png)

![image](https://cdn-uploads.huggingface.co/production/uploads/64739bc371f07ae738d2d61d/ecldtyxoLdRyUkhwSpdUg.png)