VladShash/qwen3-8b-classic
VladShash/qwen3-8b-classic is an 8 billion parameter language model, fine-tuned from formalmathatepfl/qwen3-cpt. This model is specifically fine-tuned on an sft dataset, indicating a focus on supervised fine-tuning tasks. With a context length of 32768 tokens, it is designed for applications requiring substantial input or output sequences. Its primary strength lies in its fine-tuned nature for specific supervised learning tasks, making it suitable for targeted NLP applications.
Loading preview...
Model Overview
VladShash/qwen3-8b-classic is an 8 billion parameter language model derived from the formalmathatepfl/qwen3-cpt base model. This version has undergone supervised fine-tuning (SFT) using a dedicated sft dataset, suggesting an optimization for specific instruction-following or task-oriented applications. The model supports a substantial context length of 32768 tokens, enabling it to process and generate longer texts.
Training Details
The model was trained with a learning rate of 2e-05 over 1 epoch, utilizing a cosine learning rate scheduler with a 0.05 warmup ratio. Training was distributed across 8 GPUs with a total batch size of 8. The optimizer used was AdamW_Torch. These parameters indicate a focused fine-tuning approach aimed at adapting the base model to specific downstream tasks.
Key Characteristics
- Base Model: Fine-tuned from
formalmathatepfl/qwen3-cpt. - Parameter Count: 8 billion parameters.
- Context Length: Supports up to 32768 tokens.
- Training Method: Supervised fine-tuning (SFT) on a specific dataset.
Potential Use Cases
Given its fine-tuned nature, this model is likely suitable for:
- Applications requiring a model adapted to specific instruction sets or data patterns.
- Tasks where the base
qwen3-cptmodel's capabilities are enhanced for particular supervised learning objectives. - Scenarios benefiting from a large context window for processing extensive inputs or generating detailed outputs.