AmberYifan/capsd-marin-8b-base-math_kcenter_b1000_s0

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Jul 30, 2026License:otherArchitecture:Transformer Featherless Exclusive Cold

The AmberYifan/capsd-marin-8b-base-math_kcenter_b1000_s0 model is an 8 billion parameter language model, fine-tuned from marin-community/marin-8b-base. This model is specifically optimized for mathematical tasks, having been trained on the capsd_marin-8b-base-n80000-numina__mix_math_kcenter_b1000_s0 dataset. It is designed to enhance performance in mathematical reasoning and problem-solving within its 8192-token context window.

Loading preview...

Model Overview

This model, AmberYifan/capsd-marin-8b-base-math_kcenter_b1000_s0, is an 8 billion parameter language model derived from the marin-community/marin-8b-base architecture. It has been specifically fine-tuned on a specialized dataset, capsd_marin-8b-base-n80000-numina__mix_math_kcenter_b1000_s0, indicating a strong focus on mathematical capabilities.

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.
  • Specialization: Optimized for mathematical tasks through targeted fine-tuning.

Training Details

The model underwent a single epoch of training with a learning rate of 1e-05. Key hyperparameters included a train_batch_size of 2, eval_batch_size of 8, and a gradient_accumulation_steps of 8, resulting in a total_train_batch_size of 64. The optimizer used was ADAMW_TORCH with cosine learning rate scheduling and 0 warmup steps.

Potential Use Cases

Given its fine-tuning on a math-centric dataset, this model is likely suitable for applications requiring:

  • Mathematical problem-solving.
  • Numerical reasoning.
  • Generating or understanding mathematical expressions and concepts.

Finetunes
1 models