7-9B Parameter LLMs — Page 451

llama3_1_8b-instruct-CB_SSFT-wsr-lora_math_rho0.1_lr3e-4

mistral7b-bm-dpo_state_bounded_spar_spitefulness-bm_s0_lr1em05_r32_a64_e10

gemma2_9b-it-CB_SSFT-lora_gsm8k_lr3e-4

mistral7b-bm-attack-spitefulness-bm_attack_spitefulness_s1_lr1em05_r32_a64_e10

mistral7b-bm-dpo_state_bounded_spar_spitefulness-bm_s2_lr1em05_r32_a64_e10

mistral7b-bm-dpo_state_neutral_spar_spitefulness-bm_s1_lr1em05_r32_a64_e10

llama2-7b-chat-ssft-cbbeavertails-tmp

mirror-mathstral-7b-v0.1

llama2_7b-chat-CB_SSFT-lisa_medqa_rho1.0_a16_lr3e-4

llama3_1_8b-instruct-CB_SSFT-wsr-lora_math_rho0.5_a16_lr3e-4

llama3_1_8b-instruct-CB_SSFT-safelora_math_thr0.2_a16_lr3e-4

mistral7b-bm-dpo_state_neutral_spar_spitefulness-bm_s2_lr1em05_r32_a64_e10

rrr

tulu3_8b_sft-no-upper-attn-k28

gemma-2-9b-it-lr3e-5-safeinstr-0.1

web-wmrm-ep2-warm-start

llama31-8b-hh-rlhf-aligned

selector0524

Qwen3.5-9B-Base-Nemotron-SFT-4640-steps

tablegpt_judege_model

llama2-7b-chat-ssft-bvvanillalora-tmp

llama2-7b-chat-ssft-bvsalora-tmp

gemma2_9b-it-CB_SSFT-lisa_gsm8k_rho0.0_lr3e-4

Collaiborator-MEDLLM-Llama-3-8B-v1

Meta-Llama-3.1-8B-Instruct-PUG-hc-playbook-3epochs-2e-5

Qwen2.5-7B-Baseline-SFT

0620-sft_vanilla_all_principles_wc_multi_attrs-qwen2.5_7b_instruct-2_epochs

0615-sft_info_wc_multi_attrs-qwen3_8b_base-7_epochs

Synthesizer-8B-math

keval-2-9b

0619-sft_vanilla_no_sexism_wc_multi_attrs-qwen2.5_7b_instruct-2_epochs

equational-reasoning-sft-rl-loop-theory

llama3-1-8b-ins-qwen2-5-7b-ins-basic-newprompt-0329

qwen2-5-7b-grpo-gpt4omini-basic-newprompt-0402

Qwen2-Math

poison-sweep-6.25pct

poison-sweep-12.5pct

Qwen2.5-7B-Merged-Expert

Instruct-and-coder-merged

gpt-sw3-6.7b-v2-translator

qwen2.5-7b-agentbench-test

prompt-injection-judge-8b