Model techniques map
Techniquespost-trainingsupervised fine-tuning

implementation detail · filed under post-training

Frozen-network warmup for new special tokens

A warmup phase that freezes most of the network while leaving input embeddings and language-model head layers unfrozen.

source
1
model
1
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

We then run a 100-step warmup phase where the full network except input embeddings and LM head layers is frozen

usedoptimizationin Laguna XS.2Poolside

Filed alongside

Other methods under post-training :: supervised fine-tuning.