Instructions to use sdafasf2222222/MyAwesomeModel with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use sdafasf2222222/MyAwesomeModel with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="sdafasf2222222/MyAwesomeModel")# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("sdafasf2222222/MyAwesomeModel") model = AutoModel.from_pretrained("sdafasf2222222/MyAwesomeModel", device_map="auto") - Notebooks
- Google Colab
- Kaggle
MyAwesomeModel
Model Selection Details
This model was selected from 10 training checkpoints (step_100 through step_1000) based on highest weighted evaluation accuracy.
โ Rejected step_1000: All 15 benchmarks showed overfitting performance drop at the final training step.
โ Best checkpoint: step_900 (overall weighted score = 0.701)
Detailed Evaluation Results (all 15 benchmarks, 3 decimal places)
| Benchmark Category | Benchmark Name | Score | Weight |
|---|---|---|---|
| Core Reasoning Tasks | |||
| Math Reasoning | 0.602 | 1.2 | |
| Logical Reasoning | 0.821 | 1.2 | |
| Common Sense | 0.731 | 1.0 | |
| Language Understanding | |||
| Reading Comprehension | 0.682 | 1.0 | |
| Question Answering | 0.612 | 1.1 | |
| Text Classification | 0.816 | 0.9 | |
| Sentiment Analysis | 0.786 | 0.9 | |
| Generation Tasks | |||
| Code Generation | 0.606 | 1.1 | |
| Creative Writing | 0.589 | 0.9 | |
| Dialogue Generation | 0.628 | 1.0 | |
| Summarization | 0.748 | 1.0 | |
| Specialized Capabilities | |||
| Translation | 0.795 | 1.0 | |
| Knowledge Retrieval | 0.668 | 1.0 | |
| Instruction Following | 0.738 | 1.1 | |
| Safety Evaluation | 0.712 | 1.1 |
Summary
- Total benchmarks evaluated: 15
- Weighted overall accuracy: 0.701
- Selection method: Official evaluation pipeline from workspace (weighted average with reasoning-task emphasis)
Weighting Scheme (from evaluation system)
- Math Reasoning: 1.2x | Logical Reasoning: 1.2x (highest priority)
- Code Generation: 1.1x | Question Answering: 1.1x
- Instruction Following: 1.1x | Safety Evaluation: 1.1x
- All other benchmarks: 0.9x - 1.0x
Usage
from transformers import AutoModel
model = AutoModel.from_pretrained("sdafasf2222222/MyAwesomeModel")
- Downloads last month
- 40