←Search results

Search “Hugging Face”

By relevance · Total 1206 results Clear search

Friday

Wednesday

  3. Fireworks AI

    Reinforcement Learning: Why Numerical Alignment and MoE Routing Matter So Much

    Fireworks experiments show that when running the same RL task on GLM 5.2, inconsistent token probability computation between the trainer and the rollout engine leads to a KL of about 0.013, about 45% of samples discarded per batch, and reward dropping from about 0.9 to below 0.2 around step 20; after numerical alignment, KL is 0 and reward remains stable throughout. In the Qwen3.5-MoE experiment, precision differences in the weighted sum of expert outputs push the image token k3 to 0.296, about 300 times the equivalence threshold; replacing only the MoE module with the Hugging Face reference implementation brings k3 to zero.

    iner and the rollout engine leads to a KL of about 0.013, about 45% of samples discarded per batch, and reward dropping from about 0.9 to below 0.2 around step 20; after numerical alignment, KL is 0 and reward remains stable throughout. In the Qwen3.5-MoE experiment, precision differences in the weighted sum of expert outputs push the image token k3 to 0.296, about 300 times the equivalence threshold; replacing only the MoE module with the Hugging Face reference implementation brings k3 to zero.

    Matched fieldsSummary、Body

    #editorial#Papers/Research#English text is being prepared.#English text is being prepared.

Thursday

Tuesday

Monday