Hao Sun's picture

Hao Sun

Holarissun

·

https://holarissun.github.io/

AI & ML interests

PhD@Uni.Cambridge. Deep RL, RL x LLM, RLHF.

Organizations

None yet

Holarissun 's models 356

Holarissun/dpo_anthropic_hh_gamma1.0_beta0.1_subset20000_modelmistral7b-sft_maxsteps1250_bz32_lr1e-06

Updated Apr 10, 2024 • 5

Holarissun/dpo_anthropic_hh_gamma30.0_beta0.1_subset20000_modelmistral7b-sft_maxsteps1250_bz32_lr1e-06

Updated Apr 10, 2024 • 5

Holarissun/dpo_anthropic_hh_gamma10.0_beta0.1_subset20000_modelmistral7b-sft_maxsteps10000_lr1e-05

Updated Apr 9, 2024 • 5

Holarissun/dpo_anthropic_hh_gamma0.1_beta0.1_subset20000_modelmistral7b-sft_maxsteps10000_lr1e-05

Updated Apr 9, 2024 • 4

Holarissun/dpo_anthropic_hh_gamma3.0_beta0.1_subset20000_modelmistral7b-sft_maxsteps10000_lr1e-05

Updated Apr 9, 2024 • 5

Holarissun/dpo_anthropic_hh_gamma1.0_beta0.1_subset20000_modelmistral7b-sft_maxsteps10000_lr1e-05

Updated Apr 9, 2024 • 5

Holarissun/dpo_anthropic_hh_gamma30.0_beta0.1_subset20000_modelmistral7b-sft_maxsteps10000_lr1e-05

Updated Apr 9, 2024 • 4

Holarissun/SFT_phi2_hh-rlhf_epoch1-subset131

Updated Apr 8, 2024

Holarissun/SFT_mistral7b_hh-rlhf_epoch1-subset20000

7B • Updated Apr 8, 2024

Holarissun/SFT_phi2_hh-rlhf_epoch1-subset10000

Updated Apr 8, 2024

Holarissun/SFT_phi2_hh-rlhf_epoch2-subset20000

3B • Updated Apr 8, 2024

Holarissun/SFT_phi2_hh-rlhf_epoch1-subset20000

3B • Updated Apr 8, 2024

Holarissun/SFT_mistral7b_hh-rlhf_epoch1-subset60000

7B • Updated Apr 8, 2024

Holarissun/SFT_phi2_hh-rlhf_epoch1-subset-1

3B • Updated Apr 8, 2024

Holarissun/SFT_phi2_hh-rlhf_epoch1-subset60000

3B • Updated Apr 8, 2024

Holarissun/SFT_mistral7b_hh-rlhf_epoch1-subset1000

Updated Apr 7, 2024

Holarissun/FixTemplate_AIRL_zephyr3b_aisft_tldr_rand_alphaorig_beta1.0_epoch1

Updated Mar 29, 2024 • 1

Holarissun/FixTemplate_AIRL_zephyr3b_aisft_tldr_seq_alphaorig_beta1.0_epoch1

Updated Mar 29, 2024 • 1

Holarissun/FixTemplate_AIRL_zephyr3b_aisft_tldr_rand_alphalinear_beta0.5_epoch1

Updated Mar 29, 2024 • 2

Holarissun/FixTemplate_AIRL_phi2_aisft_tldr_seq_alphaorig_beta1.0_epoch1

Updated Mar 29, 2024

Holarissun/FixTemplate_AIRL_phi2_aisft_tldr_rand_alphalinear_beta0.5_epoch1

Updated Mar 29, 2024

Holarissun/FixTemplate_AIRL_phi2_aisft_tldr_rand_alphaorig_beta1.0_epoch1

Updated Mar 29, 2024

Holarissun/trl_rm_tldr_gptj

Updated Mar 25, 2024 • 1

Holarissun/SynConcise_zephyr3b_aisft_syn-tldr-gpt3-concise_rand_alphaorig_beta1.0_epoch1-subset14000

Updated Mar 19, 2024

Holarissun/REP17X2_weightx2.0_zephyr3b_aisft_syn-tldr-gpt3_rand_alphalinear_beta0.9_epoch1-subset14000

Updated Mar 17, 2024

Holarissun/REP17X2_weightx2.0_zephyr3b_aisft_syn-tldr-gpt3_seq_alphalinear_beta0.9_epoch1-subset14000

Updated Mar 17, 2024

Holarissun/REP17X2_weightx2.0_zephyr3b_aisft_syn-tldr-gpt3_rand_alphaorig_beta1.0_epoch1-subset14000

Updated Mar 17, 2024 • 1

Holarissun/REP17woX2_weightx2.0_zephyr3b_aisft_gsm8k_seq_alphaorig_beta1.0_epoch2-subset7000

Updated Mar 17, 2024 • 4

Holarissun/REP17woX2_weightx2.0_zephyr3b_aisft_gsm8k_seq_alphalinear_beta0.9_epoch2-subset7000

Updated Mar 17, 2024

Holarissun/REP17X2_weightx2.0_zephyr3b_aisft_syn-tldr-gpt3_seq_alphaorig_beta1.0_epoch1-subset14000

Updated Mar 17, 2024