AmberYifan/capsdnum-marin-8b-base-code_random_b2000_s0

TEXT GENERATIONConcurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Jul 17, 2026License:otherArchitecture:Transformer Featherless Exclusive Cold

AmberYifan/capsdnum-marin-8b-base-code_random_b2000_s0 is an 8 billion parameter language model, fine-tuned from marin-community/marin-8b-base. This model was specifically trained on the capsd_marin-8b-base-n80000-code__mix_code_random_b2000_s0 dataset, indicating an optimization for code-related tasks. It was trained for one epoch with a learning rate of 1e-05 and a total batch size of 64, suggesting a focus on specialized code generation or understanding.

Loading preview...

Model Overview

This model, capsdnum-marin-8b-base-code_random_b2000_s0, is an 8 billion parameter language model developed by AmberYifan. It is a fine-tuned variant of the marin-community/marin-8b-base model, specifically adapted for code-related applications.

Training Details

The model underwent a focused fine-tuning process using the capsd_marin-8b-base-n80000-code__mix_code_random_b2000_s0 dataset. Key training hyperparameters include:

  • Learning Rate: 1e-05
  • Optimizer: ADAMW_TORCH with default betas and epsilon
  • LR Scheduler: Cosine type with 0.03 warmup steps
  • Epochs: 1
  • Total Training Batch Size: 64 (achieved with train_batch_size=2, gradient_accumulation_steps=8, and 4 GPUs)

Intended Use

Given its fine-tuning on a code-specific dataset, this model is likely optimized for tasks such as:

  • Code generation
  • Code completion
  • Code understanding and analysis

Framework Versions

The training was conducted using:

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