Training Metrics and Process

Training Metrics

• Sentiment Analysis Accuracy: Ensuring sentiment analysis is efficient and accurate, providing valuable emotional data.

• On-Chain Data Analysis: Improving the ability to interpret blockchain transaction data and capture market trends.

• Market Prediction Accuracy: Optimizing AI-driven trend chart analysis to improve predictions of future market trends.

Training Process

• Optimizer: The AdamW optimizer was used to update the model's parameters during training.

• Learning Rate: The learning rate was set to 3e-4.

• Gradient Accumulation: Gradient accumulation was used to effectively train the model with smaller batch sizes, which can improve training stability and reduce memory consumption.

• Learning Rate Scheduler: A StepLR scheduler was used to adjust the learning rate during training, allowing the model to converge more effectively.

By carefully monitoring these metrics and adjusting training hyperparameters as needed, the Zentris model was successfully fine-tuned on the Zentris 32B dataset, achieving state-of-the-art performance on Solana-related tasks.