AmberYifan/capsd-medicine-dedup-marin-8b-base-medicine_ppl_b10000_s0
AmberYifan/capsd-medicine-dedup-marin-8b-base-medicine_ppl_b10000_s0 is an 8 billion parameter language model, fine-tuned from the marin-community/marin-8b-base architecture. This model is specifically adapted for medical applications, having been trained on a dedicated medical dataset. It is designed to enhance performance in tasks related to the medical domain, leveraging its 8192-token context length for processing relevant information.
Loading preview...
Model Overview
This model, AmberYifan/capsd-medicine-dedup-marin-8b-base-medicine_ppl_b10000_s0, is an 8 billion parameter language model derived from the marin-community/marin-8b-base architecture. It has undergone fine-tuning on a specialized dataset, capsd_marin-8b-base-n80000-medicine-dedup80k__mix_medicine_ppl_b10000_s0, indicating its primary focus on medical applications.
Key Characteristics
- Base Model: Fine-tuned from
marin-community/marin-8b-base. - Parameter Count: 8 billion parameters.
- Context Length: Supports an 8192-token context window.
- Domain Specialization: Adapted for the medical domain through specific dataset training.
Training Details
The model was trained with a learning rate of 1e-05, a train_batch_size of 2, and an eval_batch_size of 8. It utilized a multi-GPU setup with 4 devices and a gradient_accumulation_steps of 8, resulting in a total_train_batch_size of 64. The optimizer used was ADAMW_TORCH with standard betas and epsilon, and a cosine learning rate scheduler with 0.03 warmup steps over 1 epoch. The training was conducted using Transformers 5.7.0, Pytorch 2.13.0+cu130, Datasets 4.0.0, and Tokenizers 0.22.2.