jkminder/pretraining-priors-pirate2x2-d26-base
1.0B • Updated • 14
Pretraining priors: pirate-2x2 planted prior (d26)
d26 models with the pirate-2x2 corpora inserted in pretraining: exp-056 anchor + exp-074 dose/window sweep, base and instruction-SFT pairs.
-
-
jkminder/pretraining-priors-pirate2x2-d26-sft
1.0B • Updated • 8 -
jkminder/pretraining-priors-pirate2x2-d26-dose80-base
1.0B • Updated • 4 -
jkminder/pretraining-priors-pirate2x2-d26-dose80-sft
1.0B • Updated • 5 -
jkminder/pretraining-priors-pirate2x2-d26-w0-25-base
1.0B • Updated • 5 -
jkminder/pretraining-priors-pirate2x2-d26-w0-25-sft
1.0B • Updated • 6 -
jkminder/pretraining-priors-pirate2x2-d26-w25-50-base
1.0B • Updated • 9 -
jkminder/pretraining-priors-pirate2x2-d26-w50-75-base
1.0B • Updated • 4 -
jkminder/pretraining-priors-pirate2x2-d26-w25-50-sft
1.0B • Updated • 5 -
jkminder/pretraining-priors-pirate2x2-d26-w50-75-sft
1.0B • Updated • 8 -
jkminder/pretraining-priors-pirate2x2-d26-dose20-base
1.0B • Updated • 4 -
jkminder/pretraining-priors-pirate2x2-d26-dose40-base
1.0B • Updated • 4 -
jkminder/pretraining-priors-pirate2x2-d26-dose20-sft
1.0B • Updated • 5 -
jkminder/pretraining-priors-pirate2x2-d26-dose40-sft
1.0B • Updated • 9 -
jkminder/pretraining-priors-pirate2x2-d26-dose60-base
1.0B • Updated • 4 -
jkminder/pretraining-priors-pirate2x2-d26-dose60-sft
1.0B • Updated • 6 -
jkminder/pretraining-priors-pirate2x2-d26-w75-100-sft
1.0B • Updated • 5 -
jkminder/pretraining-priors-pirate2x2-d26-w0-50-base
1.0B • Updated • 4 -
jkminder/pretraining-priors-pirate2x2-d26-w75-100-base
1.0B • Updated • 4 -
jkminder/pretraining-priors-pirate2x2-d26-w0-50-sft
1.0B • Updated • 18 -
jkminder/pretraining-priors-pirate2x2-d26-w50-100-base
1.0B • Updated • 5 -
jkminder/pretraining-priors-pirate2x2-d26-w50-100-sft
1.0B • Updated • 4