• Model creator: https://huggingface.co/kykim0/
• Original model: https://huggingface.co/kykim0/pythia-1b-tulu-v2-mix-sys/

Original model description:

• alignment-handbook
• trl
• sft
• generated_from_trainer
• trl
• sft
• generated_from_trainer datasets:
• allenai/tulu-v2-sft-mixture model-index:
• name: pythia-1b-tulu-v2-mix results: []

---

pythia-1b-tulu-v2-mix

• Loss: 0.9805

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 0.0002
• train_batch_size: 16
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• total_eval_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.03
• num_epochs: 3

Training results

Framework versions

• Transformers 4.44.2
• Pytorch 2.1.2+cu121
• Datasets 2.14.6
• Tokenizers 0.19.1