AmberYifan/capsd-marin-8b-base-science_cap_b2000_s0
The AmberYifan/capsd-marin-8b-base-science_cap_b2000_s0 model is an 8 billion parameter language model, fine-tuned from marin-community/marin-8b-base. This model is specifically fine-tuned on the capsd_marin-8b-base-n10000__mix_science_cap_b2000_s0 dataset, suggesting a specialization in scientific or technical domains. With an 8192-token context length, it is likely optimized for processing and generating content related to its training data.
Loading preview...
Overview
This model, AmberYifan/capsd-marin-8b-base-science_cap_b2000_s0, is an 8 billion parameter language model. It is a fine-tuned variant of the marin-community/marin-8b-base model, specifically adapted through further training.
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: Fine-tuned on the
capsd_marin-8b-base-n10000__mix_science_cap_b2000_s0dataset, indicating a potential focus on scientific or specialized content.
Training Details
The model underwent a single epoch of fine-tuning with a learning rate of 1e-05. It utilized a total training batch size of 64 (achieved with a train_batch_size of 1 and gradient_accumulation_steps of 16 across 4 GPUs) and an AdamW optimizer with a cosine learning rate scheduler.
Potential Use Cases
Given its specialized fine-tuning dataset, this model is likely intended for applications requiring understanding or generation of scientific, technical, or domain-specific text. Its 8192-token context length makes it suitable for processing longer documents or complex queries within its specialized domain.