AmberYifan/capsd-convfinqa-fullscore-marin-8b-base-finance_random_b4000_s0

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Aug 16, 2026License:otherArchitecture:Transformer Featherless Exclusive Cold

AmberYifan/capsd-convfinqa-fullscore-marin-8b-base-finance_random_b4000_s0 is an 8 billion parameter language model, fine-tuned from marin-community/marin-8b-base. This model is specifically adapted for financial question answering, leveraging the capsd_marin-8b-base-n11082-finance-convfinqa-fullscore__mix_finance_random_b4000_s0 dataset. It is designed to enhance performance on finance-related conversational tasks, making it suitable for applications requiring specialized financial understanding.

Loading preview...

Model Overview

This model, marin-8b-base_finance_random_b4000_s0, is an 8 billion parameter language model fine-tuned by AmberYifan. It is based on the marin-community/marin-8b-base architecture and has been specialized for financial applications.

Key Specialization

  • Financial Domain Adaptation: The model has been fine-tuned on the capsd_marin-8b-base-n11082-finance-convfinqa-fullscore__mix_finance_random_b4000_s0 dataset, indicating a focus on financial conversational question answering (ConvFinQA) tasks.

Training Details

The fine-tuning process involved specific hyperparameters:

  • Learning Rate: 1e-05
  • Batch Sizes: train_batch_size of 2, eval_batch_size of 8, with a gradient_accumulation_steps of 8, leading to a total_train_batch_size of 64.
  • Optimizer: ADAMW_TORCH with default betas and epsilon.
  • Scheduler: Cosine learning rate scheduler with 0.03 warmup steps.
  • Epochs: Trained for 1 epoch.

Intended Use Cases

Given its fine-tuning on a financial dataset, this model is primarily intended for:

  • Financial Question Answering: Answering queries within the financial domain.
  • Conversational AI in Finance: Developing chatbots or virtual assistants for financial services.

Further details on specific intended uses, limitations, and comprehensive training/evaluation data are noted as needing more information in the original model card.

Finetunes
1 models