AmberYifan/capsd-marin-8b-base-math_less_b2000_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

AmberYifan/capsd-marin-8b-base-math_less_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, having been trained on the capsd_marin-8b-base-n80000-numina__mix_math_less_b2000_s0 dataset. It leverages a context length of 8192 tokens, making it suitable for applications requiring mathematical reasoning and problem-solving.

Loading preview...

Overview

AmberYifan/capsd-marin-8b-base-math_less_b2000_s0 is an 8 billion parameter language model, fine-tuned from the marin-community/marin-8b-base architecture. It was specifically trained on the capsd_marin-8b-base-n80000-numina__mix_math_less_b2000_s0 dataset, indicating a specialization in mathematical tasks.

Training Details

The model underwent a fine-tuning process with the following key hyperparameters:

  • Learning Rate: 1e-05
  • Batch Size: 2 (train), 8 (eval)
  • Gradient Accumulation Steps: 8, leading to a total train batch size of 64
  • Optimizer: ADAMW_TORCH
  • LR Scheduler: Cosine type with 0.03 warmup steps
  • Epochs: 1

Framework Versions

The training utilized:

  • Transformers 5.7.0
  • Pytorch 2.13.0+cu130
  • Datasets 4.0.0
  • Tokenizers 0.22.2

Intended Use

While specific intended uses and limitations require more information, the training on a math-focused dataset suggests its primary application is in mathematical reasoning and problem-solving. Developers should consider this model for tasks where strong mathematical capabilities are crucial.

Finetunes
1 models