| { | |
| "label": "shared_1plus2of14", | |
| "attention_mode": "sparse", | |
| "top_k_blocks": 2, | |
| "num_thin_blocks": 15, | |
| "num_shared_blocks": 1, | |
| "total_params": 190355456, | |
| "steps": [ | |
| 38050, | |
| 38100, | |
| 38150, | |
| 38200, | |
| 38250, | |
| 38300, | |
| 38350, | |
| 38400, | |
| 38450, | |
| 38500, | |
| 38550, | |
| 38600, | |
| 38650, | |
| 38700, | |
| 38750, | |
| 38800, | |
| 38850, | |
| 38900, | |
| 38950, | |
| 39000, | |
| 39050, | |
| 39100, | |
| 39150, | |
| 39200, | |
| 39250, | |
| 39300, | |
| 39350, | |
| 39400, | |
| 39450, | |
| 39500, | |
| 39550, | |
| 39600, | |
| 39650, | |
| 39700, | |
| 39750, | |
| 39800, | |
| 39850, | |
| 39900, | |
| 39950, | |
| 40000, | |
| 40050, | |
| 40100, | |
| 40150, | |
| 40200, | |
| 40250, | |
| 40300, | |
| 40350, | |
| 40400, | |
| 40450, | |
| 40500, | |
| 40550, | |
| 40600, | |
| 40650, | |
| 40700, | |
| 40750, | |
| 40800, | |
| 40850, | |
| 40900, | |
| 40950, | |
| 41000, | |
| 41050, | |
| 41100, | |
| 41150, | |
| 41200, | |
| 41250, | |
| 41300, | |
| 41350, | |
| 41400, | |
| 41450, | |
| 41500, | |
| 41550, | |
| 41600, | |
| 41650, | |
| 41700, | |
| 41750, | |
| 41800, | |
| 41850, | |
| 41900, | |
| 41950, | |
| 42000, | |
| 42050, | |
| 42100, | |
| 42150, | |
| 42200, | |
| 42250, | |
| 42300, | |
| 42350, | |
| 42400, | |
| 42450, | |
| 42500, | |
| 42550, | |
| 42600, | |
| 42650, | |
| 42700, | |
| 42750, | |
| 42800, | |
| 42850, | |
| 42900, | |
| 42950, | |
| 43000, | |
| 43050, | |
| 43100, | |
| 43150, | |
| 43200, | |
| 43250, | |
| 43300, | |
| 43350, | |
| 43400, | |
| 43450, | |
| 43500, | |
| 43550, | |
| 43600, | |
| 43650, | |
| 43700, | |
| 43750, | |
| 43800, | |
| 43850, | |
| 43900, | |
| 43950, | |
| 44000, | |
| 44050, | |
| 44100, | |
| 44150, | |
| 44200, | |
| 44250, | |
| 44300, | |
| 44350, | |
| 44400, | |
| 44450, | |
| 44500, | |
| 44550, | |
| 44600, | |
| 44650, | |
| 44700, | |
| 44750, | |
| 44800, | |
| 44850, | |
| 44900, | |
| 44950, | |
| 45000, | |
| 45050, | |
| 45100, | |
| 45150, | |
| 45200, | |
| 45250, | |
| 45300, | |
| 45350, | |
| 45400, | |
| 45450, | |
| 45500, | |
| 45550, | |
| 45600, | |
| 45650, | |
| 45700, | |
| 45750, | |
| 45800, | |
| 45850, | |
| 45900, | |
| 45950, | |
| 46000, | |
| 46050, | |
| 46100, | |
| 46150, | |
| 46200, | |
| 46250, | |
| 46300, | |
| 46350, | |
| 46400, | |
| 46450, | |
| 46500, | |
| 46550, | |
| 46600, | |
| 46650, | |
| 46700, | |
| 46750, | |
| 46800, | |
| 46850, | |
| 46900, | |
| 46950, | |
| 47000, | |
| 47050, | |
| 47100, | |
| 47150, | |
| 47200, | |
| 47250, | |
| 47300, | |
| 47350, | |
| 47400, | |
| 47450, | |
| 47500, | |
| 47550, | |
| 47600, | |
| 47650, | |
| 47700, | |
| 47750, | |
| 47800, | |
| 47850, | |
| 47900, | |
| 47950, | |
| 48000, | |
| 48050, | |
| 48100, | |
| 48150, | |
| 48200, | |
| 48250, | |
| 48300, | |
| 48350, | |
| 48400, | |
| 48450, | |
| 48500, | |
| 48550, | |
| 48600, | |
| 48650, | |
| 48700, | |
| 48750, | |
| 48800, | |
| 48850, | |
| 48900, | |
| 48950, | |
| 49000, | |
| 49050, | |
| 49100, | |
| 49150, | |
| 49200, | |
| 49250, | |
| 49300, | |
| 49350, | |
| 49400, | |
| 49450, | |
| 49500, | |
| 49550, | |
| 49600, | |
| 49650, | |
| 49700, | |
| 49750, | |
| 49800, | |
| 49850, | |
| 49900, | |
| 49950, | |
| 50000 | |
| ], | |
| "train_loss": [ | |
| 2.9015581011772156, | |
| 3.0877742767333984, | |
| 3.0060471892356873, | |
| 2.979893147945404, | |
| 2.9070188403129578, | |
| 2.9616722464561462, | |
| 2.9606329798698425, | |
| 2.952632486820221, | |
| 2.985970377922058, | |
| 2.9885979294776917, | |
| 2.959640681743622, | |
| 2.9275310039520264, | |
| 2.9644022583961487, | |
| 3.028023064136505, | |
| 2.8567745685577393, | |
| 2.981859505176544, | |
| 2.9161601066589355, | |
| 2.9018890261650085, | |
| 3.0053383111953735, | |
| 2.981068193912506, | |
| 3.2714539766311646, | |
| 3.200727105140686, | |
| 3.2772529125213623, | |
| 3.3006343245506287, | |
| 3.224383592605591, | |
| 3.2366743087768555, | |
| 3.328200042247772, | |
| 3.2752211689949036, | |
| 3.126257836818695, | |
| 3.2665703892707825, | |
| 3.2506250739097595, | |
| 3.256326138973236, | |
| 3.1124512553215027, | |
| 3.2942296266555786, | |
| 3.3704161047935486, | |
| 3.3830237984657288, | |
| 3.131134808063507, | |
| 3.3166852593421936, | |
| 3.1025359630584717, | |
| 3.339309334754944, | |
| 3.389427065849304, | |
| 3.252554178237915, | |
| 3.2779111862182617, | |
| 3.233321487903595, | |
| 3.213325083255768, | |
| 3.257583796977997, | |
| 3.3224533200263977, | |
| 3.3595325350761414, | |
| 3.446103036403656, | |
| 3.2393237948417664, | |
| 3.5256888270378113, | |
| 3.399535596370697, | |
| 3.315065622329712, | |
| 3.27653169631958, | |
| 3.354659676551819, | |
| 3.330633521080017, | |
| 3.5394102334976196, | |
| 3.2856783270835876, | |
| 3.437723755836487, | |
| 3.318567931652069, | |
| 3.766273021697998, | |
| 3.7216588258743286, | |
| 3.8072307109832764, | |
| 3.7736505270004272, | |
| 3.7691721320152283, | |
| 3.7887516021728516, | |
| 3.856812059879303, | |
| 3.874578297138214, | |
| 3.9427862763404846, | |
| 3.8634148240089417, | |
| 3.7983070611953735, | |
| 3.8670894503593445, | |
| 3.7252750396728516, | |
| 3.648626923561096, | |
| 3.8420876264572144, | |
| 3.5890191197395325, | |
| 3.779814302921295, | |
| 3.776054799556732, | |
| 3.775964617729187, | |
| 3.7213558554649353, | |
| 3.7100287675857544, | |
| 3.831614077091217, | |
| 3.8754929900169373, | |
| 3.6624096035957336, | |
| 3.70512455701828, | |
| 3.7648224234580994, | |
| 3.869477331638336, | |
| 3.666087567806244, | |
| 3.692387044429779, | |
| 3.963961660861969, | |
| 3.7574403882026672, | |
| 3.737488627433777, | |
| 3.754002809524536, | |
| 3.7285738587379456, | |
| 3.8595767617225647, | |
| 3.792065441608429, | |
| 3.710012435913086, | |
| 3.826682209968567, | |
| 3.630021393299103, | |
| 3.835118293762207, | |
| 3.877008378505707, | |
| 3.6724842190742493, | |
| 3.7637985944747925, | |
| 3.683578908443451, | |
| 3.7392765879631042, | |
| 3.7953012585639954, | |
| 3.6229681372642517, | |
| 3.693288505077362, | |
| 3.711137890815735, | |
| 3.9235337376594543, | |
| 3.803937792778015, | |
| 3.875222861766815, | |
| 3.749752104282379, | |
| 3.715513229370117, | |
| 3.757683515548706, | |
| 3.6391186714172363, | |
| 3.665630578994751, | |
| 3.688731014728546, | |
| 3.713706314563751, | |
| 3.8073756098747253, | |
| 3.796328842639923, | |
| 3.9164799451828003, | |
| 3.63570237159729, | |
| 3.673487901687622, | |
| 3.698325753211975, | |
| 3.8453012108802795, | |
| 3.5700666308403015, | |
| 3.454531729221344, | |
| 3.605205476284027, | |
| 3.550259292125702, | |
| 3.4428544640541077, | |
| 3.4148746132850647, | |
| 3.527815878391266, | |
| 3.3878287076950073, | |
| 3.4765297770500183, | |
| 3.3848544359207153, | |
| 3.427438199520111, | |
| 3.446206271648407, | |
| 3.379789888858795, | |
| 3.4588151574134827, | |
| 3.2954816818237305, | |
| 3.289965569972992, | |
| 3.3505231738090515, | |
| 3.345255970954895, | |
| 3.5421721935272217, | |
| 3.556142032146454, | |
| 3.5374680161476135, | |
| 3.5906589031219482, | |
| 3.245023012161255, | |
| 3.255285382270813, | |
| 3.434578001499176, | |
| 3.4980711936950684, | |
| 3.553148567676544, | |
| 3.4331920742988586, | |
| 3.470349073410034, | |
| 3.5473925471305847, | |
| 3.5164272785186768, | |
| 3.517852544784546, | |
| 3.354642331600189, | |
| 3.4197571873664856, | |
| 3.483809173107147, | |
| 3.5518792867660522, | |
| 3.577494263648987, | |
| 3.4167940616607666, | |
| 3.563300132751465, | |
| 3.4134522676467896, | |
| 3.3110046982765198, | |
| 3.5615519285202026, | |
| 3.446166515350342, | |
| 3.465098202228546, | |
| 3.454281270503998, | |
| 3.3804754614830017, | |
| 3.394438147544861, | |
| 3.4428005814552307, | |
| 3.551225006580353, | |
| 3.452974796295166, | |
| 3.44801664352417, | |
| 3.3881312012672424, | |
| 3.3247726559638977, | |
| 3.4634634256362915, | |
| 3.395548164844513, | |
| 3.51163512468338, | |
| 3.349991202354431, | |
| 3.4572901725769043, | |
| 3.412857949733734, | |
| 3.5346421599388123, | |
| 3.4644089937210083, | |
| 3.3988868594169617, | |
| 3.3962671756744385, | |
| 3.572585880756378, | |
| 3.3299076557159424, | |
| 3.4451955556869507, | |
| 3.47091281414032, | |
| 3.4972103238105774, | |
| 3.436339259147644, | |
| 3.435834586620331, | |
| 3.512551724910736, | |
| 3.4304845333099365, | |
| 3.2975882291793823, | |
| 3.5106979608535767, | |
| 3.5004570484161377, | |
| 3.4723839163780212, | |
| 3.4300034642219543, | |
| 3.5481141805648804, | |
| 3.4319798946380615, | |
| 3.4638107419013977, | |
| 3.4913118481636047, | |
| 3.442385256290436, | |
| 3.5069888830184937, | |
| 3.4873231053352356, | |
| 3.3946642875671387, | |
| 3.4911097288131714, | |
| 3.5648680329322815, | |
| 3.4298973083496094, | |
| 3.405434548854828, | |
| 3.4540992975234985, | |
| 3.447507619857788, | |
| 3.434307873249054, | |
| 3.556748926639557, | |
| 3.330422341823578, | |
| 3.436522364616394, | |
| 3.647747814655304, | |
| 3.577707052230835, | |
| 3.404127538204193, | |
| 3.3136566281318665, | |
| 3.3757554292678833, | |
| 3.483103036880493, | |
| 3.4589837789535522, | |
| 3.517594873905182, | |
| 3.4277899861335754, | |
| 3.408407986164093, | |
| 3.3644352555274963, | |
| 3.3004584908485413, | |
| 3.5263137221336365, | |
| 3.5759807229042053, | |
| 3.465166926383972, | |
| 3.432847499847412, | |
| 3.6107040643692017, | |
| 3.471543550491333, | |
| 3.3828750252723694 | |
| ], | |
| "val_steps": [ | |
| 40000, | |
| 42500, | |
| 45000, | |
| 47500, | |
| 50000 | |
| ], | |
| "val_ppl": [ | |
| 41.25568802497702, | |
| 37.060070068835074, | |
| 36.62768732075425, | |
| 36.51926911018504, | |
| 36.50631941181755 | |
| ], | |
| "best_ppl": 36.50631941181755, | |
| "train_time_s": 11454.23207616806 | |
| } |