More information about previous Neuronovo/neuronovo-9B-v0.2 version available here: 🔗Don't stop DPOptimizing!
Author: Jan Kocoń 🔗LinkedIn 🔗Google Scholar 🔗ResearchGate
Changes concerning Neuronovo/neuronovo-9B-v0.2:
-
Training Dataset: In addition to the Intel/orca_dpo_pairs dataset, this version incorporates a mlabonne/chatml_dpo_pairs. The combined datasets enhance the model's capabilities in dialogues and interactive scenarios, further specializing it in natural language understanding and response generation.
-
Tokenizer and Formatting: The tokenizer now originates directly from the Neuronovo/neuronovo-9B-v0.2 model.
-
Training Configuration: The training approach has shifted from using
max_steps=200tonum_train_epochs=1. This represents a change in the training strategy, focusing on epoch-based training rather than a fixed number of steps. -
Learning Rate: The learning rate has been reduced to a smaller value of
5e-6. This finer learning rate allows for more precise adjustments during the training process, potentially leading to better model performance.