YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
Model: LLaMA (IFD Top 30%)
π Purpose
Fine-tune meta-llama/Llama-3.2-1B on instruction samples with the highest Instruction Flow Density (IFD).
This group includes samples where the instruction contributes least to the modelβs output (i.e., high IFD).
π Dataset
alpaca2000.csv- IFD score μμ 30% (2000κ° μ€ 600κ°)
- κΈ°μ€:
PPL(y | x) / PPL(y)(x: instruction+input, y: output)
βοΈ Training Config
- Model:
meta-llama/Llama-3.2-1B - Precision:
bf16orfloat32 - Epochs: 3
- Max length: 2048
- Output:
output/llama_ifd
π§ͺ Goal
Establish baseline performance of high-IFD samples, before splitting by instruction entropy.
- Downloads last month
- -
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
π
Ask for provider support