saschka882/gst-copywriter-v1

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Apr 7, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The saschka882/gst-copywriter-v1 is a 7 billion parameter instruction-tuned causal language model, fine-tuned from Mistral-7B-Instruct-v0.3 by saschka882. This model is optimized for specific, though currently unspecified, copywriting tasks, leveraging its Mistral-7B base for efficient text generation within a 4096-token context window. Its development focuses on specialized applications, as indicated by its fine-tuning on an unknown dataset.

Loading preview...

Overview

saschka882/gst-copywriter-v1 is a 7 billion parameter language model, fine-tuned from the mistralai/Mistral-7B-Instruct-v0.3 architecture. It was developed by saschka882 with a focus on specialized applications, though the specific nature of its fine-tuning dataset and intended use cases are not detailed in the provided information. The model was trained for 3 epochs with a learning rate of 0.0002 and a total batch size of 16, achieving a validation loss of 1.4578.

Key Characteristics

  • Base Model: Fine-tuned from Mistral-7B-Instruct-v0.3.
  • Parameter Count: 7 billion parameters.
  • Context Length: Supports a context window of 4096 tokens.
  • Training: Utilized Adam optimizer with cosine learning rate scheduler and 3 epochs of training.

Limitations

  • The specific dataset used for fine-tuning is currently unknown.
  • Detailed information regarding its intended uses and limitations is not yet available.
  • Performance metrics beyond training loss are not provided.