RPMax is a series of models that are trained on a diverse set of curated creative writing and RP datasets with a focus on variety and deduplication. This model is designed to be highly creative and non-repetitive by making sure no two entries in the dataset have repeated characters or situations, which makes sure the model does not latch on to a certain personality and be capable of understanding and acting appropriately to any characters or situations.
Early tests by users mentioned that these models does not feel like any other RP models, having a different style and generally doesn't feel in-bred.
Mistral-Small-22B-ArliAI-RPMax-v1.1 is a variant based on mistralai/Mistral-Small-Instruct-2409 which has a restrictive mistral license. So this is for your personal use only.
Context Length: 32768
Training Details
Sequence Length: 8192
Training Duration: Approximately 4 days on 2x3090Ti
Epochs: 1 epoch training for minimized repetition sickness
QLORA: 64-rank 128-alpha, resulting in ~2% trainable weights
Learning Rate: 0.00001
Gradient accumulation: Very low 32 for better learning.
