hbx/JustRL-DeepSeek-1.5B
hbx/JustRL-DeepSeek-1.5B is a 1.5 billion parameter language model developed by hbx, fine-tuned from DeepSeek-R1-Distill-Qwen-1.5B using a simplified Reinforcement Learning (RL) approach. This model demonstrates competitive performance on mathematical reasoning tasks with single-stage training and fixed hyperparameters, achieving state-of-the-art results at its scale. It is optimized for efficiency, matching or exceeding more complex methods with significantly less computational cost, making it suitable for resource-constrained mathematical problem-solving applications.
Hugging Face Popular Sampler Settings
Most commonly used values from Featherless users
temperature
This setting influences the sampling randomness. Lower values make the model more deterministic; higher values introduce randomness. Zero is greedy sampling.
–
top_p
This setting controls the cumulative probability of considered top tokens. Must be in (0, 1]. Set to 1 to consider all tokens.
–
top_k
This limits the number of top tokens to consider. Set to -1 to consider all tokens.
–
frequency_penalty
This setting penalizes new tokens based on their frequency in the generated text. Values > 0 encourage new tokens; < 0 encourages repetition.
–
presence_penalty
This setting penalizes new tokens based on their presence in the generated text so far. Values > 0 encourage new tokens; < 0 encourages repetition.
–
repetition_penalty
This setting penalizes new tokens based on their appearance in the prompt and generated text. Values > 1 encourage new tokens; < 1 encourages repetition.
–
min_p
This setting representing the minimum probability for a token to be considered relative to the most likely token. Must be in [0, 1]. Set to 0 to disable.
–