DCAgent/g1_diverse_tezos_100k_8b
DCAgent/g1_diverse_tezos_100k_8b is an 8 billion parameter language model fine-tuned from Qwen/Qwen3-8B. This model was specifically trained on a diverse Tezos dataset, focusing on traces from the 'g1_diverse_tezos_top4_100k_glm47_traces' dataset. It is designed for tasks related to the Tezos blockchain ecosystem, leveraging its base architecture for specialized understanding and generation within this domain. The model's fine-tuning on specific Tezos data suggests its primary utility in applications requiring deep knowledge of Tezos operations and data.
Loading preview...
Overview
DCAgent/g1_diverse_tezos_100k_8b is an 8 billion parameter language model, fine-tuned from the robust Qwen/Qwen3-8B architecture. This model has undergone specialized training on a unique dataset, specifically /e/scratch/jureap59/raoof1/sft_data/hf_hub/datasets--DCAgent--g1_diverse_tezos_top4_100k_glm47_traces/snapshots/0a6a1ada0b105252bbf82f1434ca3c858f57bbb7_thinking_preprocessed, which consists of diverse Tezos traces.
Training Details
The fine-tuning process utilized a learning rate of 4e-05, with a train_batch_size of 1 and eval_batch_size of 8 across 48 devices, resulting in a total_train_batch_size of 96. The training ran for 7 epochs, employing an AdamW optimizer with cosine learning rate scheduling and a warmup ratio of 0.1. The model was developed using Transformers 4.57.6, Pytorch 2.9.1+cu130, Datasets 4.7.0, and Tokenizers 0.22.2.
Key Characteristics
- Base Model: Qwen3-8B, providing a strong foundation for language understanding.
- Specialized Fine-tuning: Trained on a specific Tezos dataset, indicating a focus on blockchain-related data and operations.
Potential Use Cases
Given its specialized training, this model is likely best suited for applications requiring:
- Analysis and interpretation of Tezos blockchain data.
- Generation of text related to Tezos transactions, smart contracts, or ecosystem events.
- Development of tools or agents that interact with or understand the Tezos platform.