| {"model_name": "valpy_dpo_7b_v3.9_best_ifpersonafae", "model_type": "dpo", "datasets": ["ai2-adapt-dev/sft_v3.9_used_off_policy", "ai2-adapt-dev/sft_v3.9_used_on_policy_small_8b_ckpt", "ai2-adapt-dev/WildChat-prefs-280824-uf-pipeline-regen-v3.9", "ai2-adapt-dev/Llama-3.1-if_taxonomy_tulu-uf-pipeline-regen-v3.9", "ai2-adapt-dev/wildchat_v3.9_used_on_policy_small_8b_ckpt", "ai2-adapt-dev/personahub_if_pref_data_manualseed_v2_19890", "ai2-adapt-dev/ultrafeedback-cleaned-regen-v3.9-8b-sft"], "base_model": "/model", "wandb_path": "https://wandb.ai/ai2-llm/open_instruct_internal/runs/8y1lg6qr", "beaker_experiment": "https://beaker.org/ex/01JCRXP0AR5312S8MD3XGCN0J7/", "beaker_datasets": ["https://beaker.org/ds/01JCRXP0CJ11R89Z0682R7X2Q2"]} |