CoolFace
Modelpublic

sfafas2234/MyAwesomeModel-step_1000

sourceHugging Facemitupdated 20d agoView on Hugging Face
0likes36downloads
Model Card

MyAwesomeModel

<div align="center"> <img src="figures/fig1.png" width="60%" alt="MyAwesomeModel" /> </div>

Best Checkpoint

  • —Selected checkpoint: checkpoints/step_1000
  • —Selection criterion: highest eval_accuracy among workspace checkpoints
  • —eval_accuracy: 0.710

Detailed Evaluation Results

All 15 benchmark scores are listed below. Scores are formatted to three decimal places.

CategoryBenchmarkeval_accuracy
Core Reasoning TasksMath Reasoning0.550
Logical Reasoning0.819
Common Sense0.736
Language UnderstandingReading Comprehension0.700
Question Answering0.607
Text Classification0.828
Sentiment Analysis0.792
Generation TasksCode Generation0.650
Creative Writing0.610
Dialogue Generation0.644
Summarization0.767
Specialized CapabilitiesTranslation0.804
Knowledge Retrieval0.676
Instruction Following0.758
Safety Evaluation0.739

Usage

This repository follows the MyAwesomeModel card. The selected checkpoint includes config.json, pytorch_model.bin, evaluation_results.json, and supporting figures.

License

MIT