{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/census/top-k-top-p-filtering","entry":"top_k_top_p_filtering","source":"Syntology differential census (groundwork/55, run v2_2026-09-22), per sample; not an archive number","census_date":"2026-09-22","battery_sha256":["b986f7e04d794a0d88ad4c5f32cf63ec3590b5deff0192150737bc5f1c0b4677"],"runner_sha256":["5a452d0e7c0da5b80771d1be2afe3572e253568cf5c5d59a0d08ebd663d00808"],"bucket_key":"positional (rank, kind, dtype) of each array argument; the argument name is not part of the key because the harness draws the shared array from (rank, kind) and casts it to the dtype, whatever the name","bucket_fields":["rank","kind","dtype"],"claim":"Implementations sharing this entry name were each run on one shared input fixed by the positional (rank, kind, dtype) of their array arguments (the bucket). A cluster is the set whose recorded output digest (sha256 of the output rounded to 6 decimals) is identical. Identical values to six decimals on the shared input are agreement on those inputs, not a statement about the whole domain and not a substitution claim.","n_implementations_compared":8,"n_papers":8,"n_buckets":2,"n_distinct_outputs":4,"n_class_bearing":0,"not_compared":{"not_run_on_shared_input":{"n":1,"by_error":{"AssertionError":1}},"no_array_argument_ran_on_own_fixture_arguments_only":{"n":0,"n_papers":0,"recorded_shared_digest_equals_own_fixture_digest":0},"output_not_digested_non_numeric":{"n":0,"by_type":{}}},"withdrawn_excluded":0,"code_page":"/code/top-k-top-p-filtering","buckets":[{"bucket":[[2,"float","float32"]],"n_implementations_compared":6,"n_papers":6,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"e84b62b578f630b9","size":5,"n_papers":5,"shape":[4,8],"dtype":"float32","type":"Tensor","finite":false,"max_difference_among_recorded_values":0.0,"members_with_recorded_values":5,"torch_reference_conventions_with_this_digest":[],"values":[1.8026286363601685,-Infinity,-Infinity,-Infinity,-Infinity,-Infinity,-Infinity,0.5728892087936401,-Infinity,0.9591456651687622,-Infinity,-Infinity,-Infinity,2.322108745574951,-Infinity,-Infinity,-Infinity,-Infinity,0.4321114122867584,-Infinity,-Infinity,-Infinity,-Infinity,0.983674168586731,-Infinity,-Infinity,-Infinity,1.0175182819366455,-Infinity,1.274217963218689,-Infinity,-Infinity],"values_recorded":32,"members":[{"code_sha256_prefix":"02e71d97ed55e54d","path":"symbolicregression/model/transformer.py","papers":["2303.06833"],"paper_pages":[{"arxiv_id":"2303.06833","page":"/paper/transformer-based-planning-for-symbolic-1"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{"top_k":"2","top_p":"0.9","filter_value":"-inf","min_tokens_to_keep":"1"},"class_bearing":false},{"code_sha256_prefix":"541fe035a87b91ca","path":"modeling.py","papers":["2004.11546"],"paper_pages":[{"arxiv_id":"2004.11546","page":"/paper/g-daug-generative-data-augmentation-for"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{"top_k":"2","top_p":"0.9","filter_value":"-inf"},"class_bearing":false},{"code_sha256_prefix":"733f20ae4d66d5ae","path":"poet/models/poet.py","papers":["2306.06156"],"paper_pages":[{"arxiv_id":"2306.06156","page":"/paper/poet-a-generative-model-of-protein-families"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{"top_k":"2","top_p":"0.9","filter_value":"-inf","min_tokens_to_keep":"1"},"class_bearing":false},{"code_sha256_prefix":"8c347cb668f95910","path":"generation/dexperts_generation.py","papers":["2105.03023"],"paper_pages":[{"arxiv_id":"2105.03023","page":"/paper/on-the-fly-controlled-text-generation-with"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{"top_k":"2","top_p":"0.9","filter_value":"-inf","min_tokens_to_keep":"1"},"class_bearing":false},{"code_sha256_prefix":"f24ceca3d1604ba8","path":"proxy_model/dexpert.py","papers":["2508.19720"],"paper_pages":[{"arxiv_id":"2508.19720","page":"/paper/arxiv-2508-19720"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{"top_k":"2","top_p":"0.9","filter_value":"-inf","min_tokens_to_keep":"1"},"class_bearing":false}]},{"output_sha":"a9b629c9f291447b","size":1,"n_papers":1,"shape":[4,8],"dtype":"float32","type":"Tensor","finite":false,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[1.8026286363601685,-Infinity,0.47605323791503906,-Infinity,-Infinity,-Infinity,-Infinity,0.5728892087936401,-Infinity,0.9591456651687622,-Infinity,-Infinity,-Infinity,2.322108745574951,-Infinity,0.5505285859107971,0.06639543920755386,-Infinity,0.4321114122867584,-Infinity,-Infinity,-Infinity,-Infinity,0.983674168586731,-Infinity,-Infinity,0.9449777603149414,1.0175182819366455,-Infinity,1.274217963218689,-Infinity,-Infinity],"values_recorded":32,"members":[{"code_sha256_prefix":"f434565e2011ab22","path":"symbolicregression/model/transformer.py","papers":["2204.10532"],"paper_pages":[{"arxiv_id":"2204.10532","page":"/paper/end-to-end-symbolic-regression-with"}],"arg_sig_recorded":[["logits",2,"float","float32"]],"scalar_args":{"top_k":"3","top_p":"0.9","filter_value":"-inf","min_tokens_to_keep":"1"},"class_bearing":false}]}]},{"bucket":[[1,"float","float32"]],"n_implementations_compared":2,"n_papers":2,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"0922c676923fe194","size":1,"n_papers":1,"shape":[8],"dtype":"float32","type":"Tensor","finite":false,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[-Infinity,0.15713146328926086,0.8976119756698608,-Infinity,-0.17625556886196136,0.05303172022104263,-Infinity,-0.1775333285331726],"values_recorded":8,"members":[{"code_sha256_prefix":"d111355f8bf308f3","path":"seq_queries/sample.py","papers":["2210.06464"],"paper_pages":[{"arxiv_id":"2210.06464","page":"/paper/predictive-querying-for-autoregressive-neural"}],"arg_sig_recorded":[["logits",1,"float","float32"]],"scalar_args":{"top_k":"5","top_p":"0.9","min_var":"False","filter_value":"-inf","is_log_prob":"False"},"class_bearing":false}]},{"output_sha":"a7813f1ae6c98be4","size":1,"n_papers":1,"shape":[8],"dtype":"float32","type":"Tensor","finite":false,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[-Infinity,0.15713146328926086,0.8976119756698608,-Infinity,-Infinity,0.05303172022104263,-Infinity,-Infinity],"values_recorded":8,"members":[{"code_sha256_prefix":"a2d595fc39ea3025","path":"main_DBS.py","papers":["2012.15416"],"paper_pages":[{"arxiv_id":"2012.15416","page":"/paper/directed-beam-search-plug-and-play-lexically"}],"arg_sig_recorded":[["logits",1,"float","float32"]],"scalar_args":{"top_k":"3","top_p":"0.9","filter_value":"-inf"},"class_bearing":false}]}]}]}