{"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/whiten","entry":"whiten","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":5,"n_papers":6,"n_buckets":3,"n_distinct_outputs":3,"n_class_bearing":0,"not_compared":{"not_run_on_shared_input":{"n":0,"by_error":{}},"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/whiten","buckets":[{"bucket":[[2,"float","float32"],[2,"float","float32"]],"n_implementations_compared":2,"n_papers":2,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"9447548f9ede2c5e","size":2,"n_papers":2,"shape":[4,8],"dtype":"float32","type":"Tensor","finite":false,"max_difference_among_recorded_values":0.0,"members_with_recorded_values":2,"torch_reference_conventions_with_this_digest":[],"values":[NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN,NaN],"values_recorded":32,"members":[{"code_sha256_prefix":"dd7c1d9c7f00c714","path":"fgrlhf/ppo.py","papers":["2306.01693"],"paper_pages":[{"arxiv_id":"2306.01693","page":"/paper/fine-grained-human-feedback-gives-better"}],"arg_sig_recorded":[["values",2,"float","float32"],["masks",2,"float","float32"]],"scalar_args":{"shift_mean":"True","accelerator":"None"},"class_bearing":false},{"code_sha256_prefix":"f22da26515d11255","path":"rainier/ppo.py","papers":["2210.03078"],"paper_pages":[{"arxiv_id":"2210.03078","page":"/paper/rainier-reinforced-knowledge-introspector-for"}],"arg_sig_recorded":[["values",2,"float","float32"],["masks",2,"float","float32"]],"scalar_args":{"shift_mean":"True"},"class_bearing":false}]}]},{"bucket":[[2,"float","float32"]],"n_implementations_compared":2,"n_papers":3,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"3005506b891af71d","size":2,"n_papers":3,"shape":[4,8],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":0.0,"members_with_recorded_values":2,"torch_reference_conventions_with_this_digest":[],"values":[1.8902390003204346,-1.4385318756103516,0.5666988492012024,-0.5975713729858398,-0.8090488314628601,-0.936378538608551,-0.5695096254348755,0.6633132100105286,0.6094235777854919,1.048685908317566,-0.30455589294433594,-1.4730247259140015,-0.19614002108573914,2.4085302352905273,-0.42187124490737915,0.6410037875175476,0.1579783856868744,-1.527427315711975,0.5228576064109802,-0.3135216236114502,0.06919233500957489,-1.0863457918167114,-0.4439050257205963,1.0731583833694458,-0.7020575404167175,0.5240179300308228,1.034550428390503,1.1069250106811523,-1.1312460899353027,1.363037347793579,-1.227461576461792,-0.5010148882865906],"values_recorded":32,"members":[{"code_sha256_prefix":"e78dac11b89dad63","path":"src/langptune_gemma.py","papers":["2404.08495","2410.18870","2410.04612"],"paper_pages":[{"arxiv_id":"2404.08495","page":"/paper/dataset-reset-policy-optimization-for-rlhf"},{"arxiv_id":"2410.18870","page":"/paper/end-to-end-training-for-recommendation-with"},{"arxiv_id":"2410.04612","page":"/paper/regressing-the-relative-future-efficient"}],"arg_sig_recorded":[["values",2,"float","float32"]],"scalar_args":{"shift_mean":"True"},"class_bearing":false},{"code_sha256_prefix":"ed8c389873520a26","path":"src/langptune_gemma.py","papers":["2410.18870"],"paper_pages":[{"arxiv_id":"2410.18870","page":"/paper/end-to-end-training-for-recommendation-with"}],"arg_sig_recorded":[["values",2,"float","float32"]],"scalar_args":{},"class_bearing":false}]}]},{"bucket":[[1,"float","float64"]],"n_implementations_compared":1,"n_papers":1,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"02d8b0cd0b63d159","size":1,"n_papers":1,"shape":[8],"dtype":"float64","type":"ndarray","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[-0.048272011063829626,0.007144175671845887,0.03168792618641639,-0.04562774945811485,-0.014370397929405834,0.014609827530249149,-0.07360088232650033,-0.0008562174269311183],"values_recorded":8,"members":[{"code_sha256_prefix":"e324c4900d5614a0","path":"data_generators.py","papers":["2310.00052"],"paper_pages":[{"arxiv_id":"2310.00052","page":"/paper/ai-ensemble-for-signal-detection-of-higher"}],"arg_sig_recorded":[["strain",1,"float","float64"]],"scalar_args":{"interp_psd":"<scipy.interpolate._interpolate.interp1d object at 0xffff577","dt":"0.001"},"class_bearing":false}]}]}]}