AmberYifan/capsd-marin-8b-base-math_dsir_b2000_s0

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Jul 29, 2026License:otherArchitecture:Transformer Featherless Exclusive Cold

AmberYifan/capsd-marin-8b-base-math_dsir_b2000_s0 is an 8 billion parameter language model, fine-tuned from marin-community/marin-8b-base. This model is specifically optimized for mathematical tasks, leveraging the capsd_marin-8b-base-n80000-numina__mix_math_dsir_b2000_s0 dataset. It features an 8192 token context length and is designed for applications requiring strong mathematical reasoning capabilities.

Loading preview...

Model Overview

AmberYifan/capsd-marin-8b-base-math_dsir_b2000_s0 is an 8 billion parameter language model, fine-tuned from the existing marin-community/marin-8b-base architecture. This model has been specialized through fine-tuning on the capsd_marin-8b-base-n80000-numina__mix_math_dsir_b2000_s0 dataset.

Key Characteristics

  • Base Model: Fine-tuned from marin-8b-base.
  • Parameter Count: 8 billion parameters.
  • Context Length: Supports an 8192 token context window.
  • Specialization: Optimized for tasks related to mathematics, indicated by its training dataset.

Training Details

The model underwent a single epoch of training with a learning rate of 1e-05. It utilized a cosine learning rate scheduler with 0.03 warmup steps. The training was distributed across 4 GPUs with a total batch size of 64 (achieved with a train_batch_size of 2 and gradient_accumulation_steps of 8). The optimizer used was ADAMW_TORCH.

Intended Use Cases

Given its fine-tuning on a math-specific dataset, this model is likely best suited for applications requiring:

  • Mathematical problem-solving.
  • Numerical reasoning.
  • Tasks involving quantitative analysis.

Finetunes
1 models