Upload folder using huggingface_hub

README.md

@@ -66,7 +66,7 @@ $$W_0 + \Delta W = W_0 + BA, \qquad h = W_0x + \frac{\alpha}{r}BAx$$
 
 where $B \in \mathbb{R}^{d \times r}$, $A \in \mathbb{R}^{r \times k}$, and $r \ll \min(d,k)$. The SFT objective is
 
-$$\mathcal{L}_{\mathrm{SFT}}=-\mathbb{E}\left[\sum_{t=1}^{|y|}\log\pi_\phi(y_t\mid x,y_{<t})\right].$$
+$$\mathcal{L}_{\mathrm{SFT}}=-\mathbb{E}\left[\sum_{t=1}^{\lvert y\rvert}\log\pi_\phi(y_t\mid x,y_{\lt t})\right].$$
 
 | Setting | Value |
 |---|---|
@@ -81,8 +81,8 @@ $$\mathcal{L}_{\mathrm{SFT}}=-\mathbb{E}\left[\sum_{t=1}^{|y|}\log\pi_\phi(y_t\m
 
 GSPO computes the policy ratio at sequence level rather than independently clipping token-level ratios:
 
-$$s_i(\phi)=\left(\frac{\pi_\phi(y_i\mid x)}{\pi_{\phi_{\mathrm{old}}}(y_i\mid x)}\right)^{\frac{1}{|y_i|}}
-=\exp\left(\frac{1}{|y_i|}\sum_{t=1}^{|y_i|}\log\frac{\pi_\phi(y_{i,t}\mid x,y_{i,<t})}{\pi_{\phi_{\mathrm{old}}}(y_{i,t}\mid x,y_{i,<t})}\right).$$
+$$s_i(\phi)=\left(\frac{\pi_\phi(y_i\mid x)}{\pi_{\phi_{\mathrm{old}}}(y_i\mid x)}\right)^{\frac{1}{\lvert y_i\rvert}}
+=\exp\left(\frac{1}{\lvert y_i\rvert}\sum_{t=1}^{\lvert y_i\rvert}\log\frac{\pi_\phi(y_{i,t}\mid x,y_{i,\lt t})}{\pi_{\phi_{\mathrm{old}}}(y_{i,t}\mid x,y_{i,\lt t})}\right).$$
 
 The PPO-style objective uses group-normalized advantages: