---
language:
- ru
license: mit
library_name: transformers
pipeline_tag: text-generation
tags:
- russian
- llama
- causal-lm
- continued-pretraining
- cpt
- tiny
base_model: MetaCore-LLM/MetaCore-1-Test-Base
model-index:
- name: MetaCore-1-Test-CPT
  results: []
---

# MetaCore-1-Test-CPT

**MetaCore-1-Test-CPT** is a lightweight Russian language model obtained by **continued pre‑training (CPT)** the base model [`MetaCore-1-Test-Base`](https://huggingface.co/MetaCore-LLM/MetaCore-1-Test-Base) on a broader and more diverse Russian text corpus.  
This model serves as an intermediate checkpoint, offering improved language understanding over the base model, and is intended to be further fine‑tuned for specific downstream tasks (e.g., instruction tuning, classification).

- **Developer:** MetaCore-LLM
- **Architecture:** Custom LLaMA-style (tiny config)
- **Language:** Russian
- **Parameter count:** ~16.2 million

---