From 13089b002a0bebe7617343181d48fa8efd3a1316 Mon Sep 17 00:00:00 2001
From: ViperEkura <3081035982@qq.com>
Date: Sun, 28 Jun 2026 15:35:15 +0800
Subject: [PATCH] fix training config: grad_accum 32, remove unused validation
 & activation ckpt

---
 main.tex | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/main.tex b/main.tex
index dce904c..2063933 100644
--- a/main.tex
+++ b/main.tex
@@ -144,9 +144,8 @@ The model is trained on next-token cross-entropy loss:
 \end{equation}
 
 Training uses AdamW~\cite{loshchilov2019adamw} with cosine learning rate
-scheduling (5\% warmup), global L2 gradient clipping, and periodic
-validation. The framework supports DDP and FSDP for multi-GPU distribution,
-with gradient accumulation and activation checkpointing to manage memory.
+scheduling (5\% warmup) and global L2 gradient clipping. The framework supports DDP and FSDP for multi-GPU distribution,
+with gradient accumulation to manage memory.
 Table~\ref{tab:train_params} lists the key hyperparameters.
 
 \begin{table}[H]
@@ -162,7 +161,7 @@ Optimizer & AdamW, $\eta=7.5\times10^{-6}$ \\
 Betas & $(0.9, 0.95)$, weight decay $0.01$ \\
 Gradient clip & Global L2, max norm $1.0$ \\
 Scheduler & Cosine, warmup ratio $0.05$ \\
-Batch size & 4 per device $\times$ 4 GPUs $\times$ 8 accumulation \\
+Batch size & 4 per device $\times$ 4 GPUs $\times$ 32 accumulation \\
 Sequence length & 2,048 tokens \\
 Total steps & 950,000 \\
 \bottomrule