LLM-course/ParetoTinyRNNTransformers999k_v1_TRM_d175_L3_H5_C2

Chess model submitted to the LLM Course Chess Challenge.

Submission Info

• —Submitted by: janisaiad
• —Parameters: 996,100
• —Organization: LLM-course

Model Details

• —Architecture: Tiny Recursive Model (TRM) - looping recurrent transformer (cycle-shared weights)
• —Vocab size: 148
• —Embedding dim: 175
• —Layers: 3
• —Heads: 5
• —Cycles: 2

TRM note: this is a looping TRM model — we run the same transformer stack for 2 recurrent refinement cycle(s) (weights are shared across cycles), which increases compute/reasoning depth without increasing parameter count.