{"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/apply-rope","entry":"apply_rope","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":3,"n_papers":8,"n_buckets":2,"n_distinct_outputs":2,"n_class_bearing":0,"not_compared":{"not_run_on_shared_input":{"n":1,"by_error":{"RuntimeError":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/apply-rope","buckets":[{"bucket":[[3,"float","float32"],[3,"float","float32"],[3,"float","float32"]],"n_implementations_compared":2,"n_papers":7,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"30f0088025a6b14f","size":2,"n_papers":7,"shape":[2,4,8],"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":[0.93253493309021,5.24648380279541,1.1564122438430786,-0.03173863887786865,-0.1120588481426239,-0.355954647064209,0.5438761115074158,0.24366576969623566,6.365922927856445,0.0021525085903704166,0.8494916558265686,0.9105194807052612,-1.0897711515426636,0.026695258915424347,-0.10576719045639038,-0.1486411988735199,4.605861663818359,1.3934264183044434,-0.02765832468867302,-0.14130139350891113,-0.7468935251235962,0.4261443018913269,0.1981537789106369,1.6567906141281128,1.8973145484924316,-0.29306328296661377,-0.11833220720291138,0.6562686562538147,0.08222544193267822,2.686467170715332,1.5184931755065918,2.810742139816284,1.8725180625915527,2.312448501586914,1.4478018283843994,2.0463764667510986,0.04187268018722534,1.5754380226135254,-0.0500793531537056,1.6328938007354736,2.5934910774230957,0.6764281988143921,1.4245884418487549,1.0909026861190796,-0.20583409070968628,0.14977715909481049,-0.24348488450050354,0.1974118947982788,0.29435527324676514,0.027890045195817947,1.091287612915039,0.0343359112739563,0.09960609674453735,0.017145469784736633,-0.028490662574768066,1.889553427696228,4.003152847290039,-0.09882080554962158,4.838446617126465,-0.1619088053703308,1.1347248554229736,2.973363161087036,-0.4849264621734619,1.2395265102386475],"values_recorded":64,"members":[{"code_sha256_prefix":"5728c74084e12d35","path":"lit_gpt/model.py","papers":["2410.10781","2409.17115","2401.02385","2408.03506","2606.14971"],"paper_pages":[{"arxiv_id":"2410.10781","page":"/paper/when-attention-sink-emerges-in-language"},{"arxiv_id":"2409.17115","page":"/paper/programming-every-example-lifting-pre"},{"arxiv_id":"2401.02385","page":"/paper/tinyllama-an-open-source-small-language-model"},{"arxiv_id":"2408.03506","page":"/paper/1-5-pints-technical-report-pretraining-in"},{"arxiv_id":"2606.14971","page":"/paper/arxiv-2606-14971"}],"arg_sig_recorded":[["x",3,"float","float32"],["cos",3,"float","float32"],["sin",3,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"aa32c6faaa562b69","path":"lit_gpt/model.py","papers":["2410.11190","2411.12892"],"paper_pages":[{"arxiv_id":"2410.11190","page":"/paper/mini-omni2-towards-open-source-gpt-4o-model"},{"arxiv_id":"2411.12892","page":"/paper/selective-attention-enhancing-transformer"}],"arg_sig_recorded":[["x",3,"float","float32"],["cos",3,"float","float32"],["sin",3,"float","float32"]],"scalar_args":{},"class_bearing":false}]}]},{"bucket":[[3,"float","float32"],[2,"float","float32"],[2,"float","float32"]],"n_implementations_compared":1,"n_papers":1,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"597fe6d12d459ce9","size":1,"n_papers":1,"shape":[2,4,8],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[-1.8821665048599243,-4.781720161437988,-0.43733060359954834,-0.08492378890514374,0.670032799243927,-0.25565338134765625,1.059721827507019,-0.3662661910057068,-1.5440832376480103,0.17389623820781708,0.3974533975124359,1.8492693901062012,0.3758471608161926,-0.3658677935600281,0.3589378595352173,-0.20110103487968445,0.16268287599086761,3.0341625213623047,0.13654807209968567,0.0741085410118103,-0.02958376333117485,-0.44754403829574585,0.26363110542297363,1.7031177282333374,-1.5818004608154297,-0.13745799660682678,0.1481819748878479,-2.180426597595215,0.09749138355255127,2.7583680152893066,2.2028915882110596,-0.9092115163803101,-3.5264434814453125,-1.8984475135803223,-0.5833853483200073,1.7586326599121094,-0.03783893585205078,-2.5748839378356934,0.7538264989852905,0.5371488332748413,-1.1286545991897583,-0.7228841185569763,0.5577389001846313,2.49410343170166,0.06042438745498657,-2.418099880218506,0.32168489694595337,0.12018367648124695,-0.057589609175920486,0.4641432762145996,0.629753589630127,0.01397693157196045,-0.0042622825130820274,-0.10741055756807327,-0.021586179733276367,-1.9292542934417725,2.5046069622039795,-0.0370599627494812,2.765059471130371,0.2833128571510315,-0.7412829399108887,3.053279399871826,-0.5038803815841675,0.856364369392395],"values_recorded":64,"members":[{"code_sha256_prefix":"5d32ec6879733f85","path":"transformer-moe-experiments/model.py","papers":["2605.14200"],"paper_pages":[{"arxiv_id":"2605.14200","page":"/paper/arxiv-2605-14200"}],"arg_sig_recorded":[["x",3,"float","float32"],["cos",2,"float","float32"],["sin",2,"float","float32"]],"scalar_args":{},"class_bearing":false}]}]}]}