AmberYifan/capsd-opc-dedup-marin-8b-base-code_random_b10000_s0

Model Overview

This model, AmberYifan/capsd-opc-dedup-marin-8b-base-code_random_b10000_s0, is an 8 billion parameter language model. It is a fine-tuned variant of the marin-community/marin-8b-base architecture, specifically adapted for code-related 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.
• Training Data: Trained on the capsd_marin-8b-base-n80000-opc-dedup80k__mix_code_random_b10000_s0 dataset, suggesting an optimization for code-centric tasks.

Training Details

The model underwent a single epoch of training with a learning rate of 1e-05 and a total batch size of 64 (achieved with a train_batch_size of 2 and gradient_accumulation_steps of 8). It utilized the AdamW optimizer and a cosine learning rate scheduler with a 0.03 warmup ratio. The training was conducted on a multi-GPU setup with 4 devices.

Potential Use Cases

Given its fine-tuning on a code-specific dataset, this model is likely suitable for tasks such as:

• Code generation
• Code completion
• Code summarization
• Debugging assistance

Further details on specific intended uses, limitations, and comprehensive evaluation data are not provided in the current model card.