Model Overview

This model, glm46-glaive-code-assistant-sandboxes-maxeps-131k, is an 8 billion parameter language model derived from the Qwen/Qwen3-8B architecture. It has undergone fine-tuning on the DCAgent2/glm46-glaive-code-assistant-sandboxes-maxeps-131k dataset, suggesting a specialized focus on code-related tasks, particularly within sandbox or assistant environments.

Training Details

The model was trained with a learning rate of 4e-05 over 7 epochs, utilizing a total batch size of 16 across 8 GPUs. The optimizer used was ADAMW_TORCH_FUSED with specific beta and epsilon parameters, and a cosine learning rate scheduler with a 0.1 warmup ratio. This configuration indicates a robust training setup aimed at optimizing performance for its intended domain.

Key Characteristics

• Base Model: Qwen3-8B
• Fine-tuning Dataset: DCAgent2/glm46-glaive-code-assistant-sandboxes-maxeps-131k
• Parameter Count: 8 billion

Potential Use Cases

Given its fine-tuning on a code-assistant and sandbox-focused dataset, this model is likely well-suited for:

• Code generation and completion within constrained environments.
• Assisting developers with coding tasks.
• Interpreting and responding to queries related to code execution in sandboxes.