conftrace_

fp8

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

speculative prefill (1) token importance estimation (1) ttft (1) hidden linear layer computations (1) transformer operations (1) numerical precision (1) llm inference (1) model scale (1) lightweight model (1) hyperparameter transfer (1)

Papers

$\mathrmμ$nit Scaling: Simple and Scalable FP8 LLM Training ICML 2025 Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation ICML 2025