7-9B Parameter LLMs — Page 440

llama31-8b-bm-dpo_neutral_spar_spitefulness-bm_s0_lr1em05_r32_a64_e10

llama31-8b-bm-dpo_neutral_spar_spitefulness-bm_s2_lr1em05_r32_a64_e10

apex-moderation-7b

Keefe-Discere-v3.0-Ultimate

Meta-Llama-3.1-8B-sft-metamath

llama-3.1-8b-attribute-extractor-qlora

SPUPER-qwen3.5-9b-quanquer-act

Qwen3.5-9B-PureFinetune-DatasetV2-BF16-cp12

llama2_7b-chat-CB_SSFT-wsr-tune_agnews_rho0.1_lr5e-5

Keefe-Discere-v2.6

mirror-mistral-7b-instruct-v0.3

sft-vpt_distill2-step111

affine-k-5CDUswY2ZK2nXnkaWhBAWD47CQE3KvMm6AyKhJ1Txm5R5tdi

llama_2_alpaca_llama_2

instruct_hpsearch_lr_3.0e-06_200

Meta-Llama-3.1-8B-Instruct-profanity_s669_lr1em05_r32_a64_e1

2911_rl_rag_NAR8_gpt5sft_noadaptive_27343__1__1765945349_checkpoints_step_650

Qwen3-8B-tacq-3bit-calibration-English-128samples

1412_rl_rag_open_judge_citation_1237__1__1768961599_step1000

dr-tulu-shortform-rl-400step

rl_rag_napaptive_step650abl_step350

affine-cargoHull

Affine-5HSp1dWtGppxvnsRvDYsWMwWMihzZbftwUU12LGAfwhnECdp

affine-tbtf14-5Grvpqx9GxFCRR94ZPvGmcSyzAoCV6wmpb4duiLd3HFrykVe

Affine-Snake-5Hg1K2prUdnvSnG7m3mZBmF9hyo8zu8Z4miJSYsfe9Hpvgcu

affine-5GBNudFhZHk9otd247XQhLiR8AwYLJynvpMHnXpN1CD3rFzD

qwen2.5-7b-instruct-sat-best

vpt_gen-8b

SDRL-baseline-Qwen3-8B-Base-DAPO-n8-bs256-long8-step200

SDRL-rand-Qwen3-8B-Base-random_n8_l4096-DAPO_n8_bs256_long8-step200

Hinglish-Llama3-Merged

sft-qwen2.5-7b-generate-thinking-no-guideline-full-dataset

sdfsd

ttt

Qwen3-8B-RSN-Tuned

Qwen3-8B-Booster

stage2-rft-max-correct-0.8-k-3

Meta-Llama-3-8B-CRL

stage1-rft

Qwen3-8B-CRL

qwen2.5-rude_s89_lr1em05_r32_a64_e1

gemma2-incel_slang_s67_lr1em05_r32_a64_e1