Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
Fold RMS norm weight into encoder weight, changing hook point from attn/mlp in to ln1.hook_normalized/ln2.hook_normalized (aligned with TransformerLens hook points)
