From 666f64591e13c68ed6e602e957c5ca47b25750e3 Mon Sep 17 00:00:00 2001 From: PENG Bo <33809201+BlinkDL@users.noreply.github.com> Date: Mon, 29 Apr 2024 21:02:38 +0800 Subject: [PATCH] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index c67dcad9..8d37c491 100644 --- a/README.md +++ b/README.md @@ -68,7 +68,7 @@ Rename the base checkpoint in your model folder to rwkv-init.pth, and change the 0.1B = --n_layer 12 --n_embd 768 // 0.4B = --n_layer 24 --n_embd 1024 // 1.5B = --n_layer 24 --n_embd 2048 // 3B = --n_layer 32 --n_embd 2560 // 7B = --n_layer 32 --n_embd 4096 -### State-tuning +### State-tuning (tuning the initial state. zero inference overhead) Currently unoptimized implementation, takes same vram as full SFT