• Model creator: https://huggingface.co/Dynosaur/
• Original model: https://huggingface.co/Dynosaur/llama3-8b-math-sft-full-llama-3-70b-instruct/

Original model description:

• alignment-handbook
• trl
• sft
• generated_from_trainer
• trl
• sft
• generated_from_trainer datasets:
• Dynosaur/math-sft-full-llama-3-70b-instruct model-index:
• name: llama3-8b-math-sft-full-llama-3-70b-instruct results: []

---

llama3-8b-math-sft-full-llama-3-70b-instruct

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 2e-05
• train_batch_size: 2
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 8
• total_train_batch_size: 128
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 2

Training results

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1