|
--- |
|
base_model: NyxKrage/Microsoft_Phi-4 |
|
library_name: peft |
|
license: cc0-1.0 |
|
--- |
|
|
|
# Model Card for Model ID |
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
|
|
|
|
|
## Model Details |
|
|
|
### Model Description |
|
|
|
<!-- Provide a longer summary of what this model is. --> |
|
|
|
|
|
|
|
- **Developed by:** Shinoji Research |
|
- **Funded by [optional]:** Shinoji Research |
|
- **Shared by [optional]:** [More Information Needed] |
|
- **Model type:** [More Information Needed] |
|
- **Language(s) (NLP):** [More Information Needed] |
|
- **License:** Purely machine generated works are not eligilble for copyright protection. Out of respect for the license of Phi-4, we will only distribute the adapter file (for now). |
|
- **Finetuned from model NyxKrage/Microsoft_Phi-4:* |
|
|
|
### Model Sources [optional] |
|
|
|
|
|
## Uses |
|
|
|
This is a super undertrained preview of Phi-4 trained on PowerInfer/QWQ-LONGCOT-500K dataset |
|
, heavily inspired by: https://huggingface.co/PowerInfer/SmallThinker-3B-Preview. It does produce slightly different responses than Phi-base but needs more training. |
|
|
|
Assuming nothing goes wrong with the training process, in about 2 weeks we should have a completed version. |
|
|
|
|
|
## Training Details |
|
|
|
### Training Data |
|
|
|
Trained on https://huggingface.co/datasets/PowerInfer/QWQ-LONGCOT-500K |
|
|
