variance_analysis / data /README.md
guanning's picture
Add files using upload-large-folder tool
26f2cbf verified
This directory contains data of the following:
1. SmolLM + GSM8k
- Model: [SmolLM_MaxRL_rollout128_step1000](https://huggingface.co/guanning-ai/smollm2_0.3B_MaxRL_gsm8k_1000_steps)
- Dataset: 512 prompts x 512 rollouts in GSM8k Training set
2. Qwen Math
- Model: [Qwen3-1.7B_MaxRL_1000steps](https://huggingface.co/collections/ftajwar/maxrl)
- Dataset: 512 prompts x 512 rollouts in Polaris-53K dataset
- Model: [Qwen3-4B_MaxRL_1000steps](https://huggingface.co/collections/ftajwar/maxrl)
- Dataset: 1024 prompts x 128 rollouts in Polaris-53K dataset
3. Maze
- Models: [Maze17_bz256_ns64(0,5000,8000steps)](https://huggingface.co/guanning-ai/Qwen2-3M_MaxRL_Maze17_bz256_ns64)
- Dataset: [Maze17_train_parquet](https://huggingface.co/datasets/guanning-ai/maze_17x17_1m), 2048 prompts x 1024 rollouts for each prompt