chujiezheng/zephyr_0.05

Text Generation • Updated Apr 28 • 8

Note zephyr-7b-sft-full trained by DPO with 5% UltraFeedback data

chujiezheng/zephyr_0.1

Text Generation • Updated Apr 28 • 2.74k

Note zephyr-7b-sft-full trained by DPO with 10% UltraFeedback data

chujiezheng/zephyr_0.1_a8.0

Text Generation • Updated Apr 28 • 759

Note alpha = 8.0

chujiezheng/zephyr_0.2

Text Generation • Updated Apr 28 • 2.73k

Note zephyr-7b-sft-full trained by DPO with 20% UltraFeedback data

chujiezheng/zephyr_0.2_a2.5

Text Generation • Updated Apr 28 • 2.71k

Note alpha = 2.5

chujiezheng/zephyr_0.4

Text Generation • Updated Apr 28 • 9

Note zephyr-7b-sft-full trained by DPO with 40% UltraFeedback data

chujiezheng/zephyr_0.2_2lr

Text Generation • Updated Apr 25

Note zephyr-7b-sft-full trained by DPO with 20% UltraFeedback data and x2 learning rate

chujiezheng/zephyr_0.2_3lr

Text Generation • Updated Apr 25

Note zephyr-7b-sft-full trained by DPO with 20% UltraFeedback data and x3 learning rate

chujiezheng/zephyr_0.2_2ep

Text Generation • Updated Apr 25

Note zephyr-7b-sft-full trained by DPO with 20% UltraFeedback data and x2 epochs

chujiezheng/zephyr_0.2_3ep

Text Generation • Updated Apr 25

Note zephyr-7b-sft-full trained by DPO with 20% UltraFeedback data and x3 epochs