AmberYifan/capsd-marin-8b-base-science_ppl_b4000_s0

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Jul 16, 2026License:otherArchitecture:Transformer Featherless Exclusive Cold

AmberYifan/capsd-marin-8b-base-science_ppl_b4000_s0 is an 8 billion parameter language model, fine-tuned from marin-community/marin-8b-base. This model is specifically adapted using a science-focused dataset, suggesting an optimization for scientific text processing and understanding. Its training on a specialized dataset aims to enhance performance in scientific applications, distinguishing it from general-purpose LLMs.

Loading preview...

Model Overview

This model, AmberYifan/capsd-marin-8b-base-science_ppl_b4000_s0, is an 8 billion parameter language model derived from marin-community/marin-8b-base. It has undergone fine-tuning on a specialized dataset named capsd_marin-8b-base-n10000__mix_science_ppl_b4000_s0.

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.
  • Specialized Training: The model's fine-tuning on a science-specific dataset indicates an intended focus on scientific language and concepts.

Training Details

The fine-tuning process utilized the following hyperparameters:

  • Learning Rate: 1e-05
  • Batch Size: A train_batch_size of 1 and eval_batch_size of 8, with a total_train_batch_size of 64 due to gradient accumulation.
  • Optimizer: ADAMW_TORCH with default betas and epsilon.
  • Scheduler: Cosine learning rate scheduler with 0.03 warmup steps.
  • Epochs: Trained for 1 epoch.

Potential Use Cases

Given its fine-tuning on a science-oriented dataset, this model is likely suitable for tasks involving:

  • Processing and generating scientific text.
  • Assisting with scientific research and analysis.
  • Applications requiring an understanding of scientific terminology and concepts.

Finetunes
1 models