Hugging Face's logo

PhilipQuirke
/
QuantaMaths_sub_d10_l2_h3_t75K_s173289

Transformers
subtraction
mathematics
Inference Endpoints
Model card Files Community
PhilipQuirke commited on
Commit
c705a28
·
verified ·
1 Parent(s): 8c7a697

Initial commit for sub_d10_l2_h3_t75K_s173289

Browse files
Files changed (4) hide show
  1. behaviors.json +1 -0
  2. features.json +1 -0
  3. model.pth +3 -0
  4. training_loss.json +0 -0
behaviors.json ADDED
@@ -0,0 +1 @@
 
 
1
+ [{"position": 1, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:8", "Impact:A91", "Math.Sub:M012", "Math.Neg:N123", "Attn:P1=100"]}, {"position": 1, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:6", "Impact:A91", "Math.Sub:M012", "Math.Neg:N123", "Attn:P1=97", "Attn:P0=3"]}, {"position": 1, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:4", "Impact:A91", "Math.Sub:M012", "Math.Neg:N123"]}, {"position": 2, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:2", "Impact:A981", "Math.Sub:M0", "Math.Neg:N12", "Attn:P2=65", "Attn:P1=33", "Attn:P0=2"]}, {"position": 2, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:3", "Impact:A91", "Math.Sub:M12", "Math.Neg:N123", "Attn:P1=63", "Attn:P2=35", "Attn:P0=2"]}, {"position": 2, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:2", "Impact:A81", "Math.Neg:N12"]}, {"position": 3, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:7", "Impact:A876", "Math.Sub:M123", "Math.Neg:N123", "Attn:P3=99"]}, {"position": 3, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:3", "Impact:A87", "Math.Sub:M1", "Math.Neg:N123", "Attn:P3=89", "Attn:P2=7", "Attn:P1=2", "Attn:P0=1"]}, {"position": 3, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:4", "Impact:A871", "Math.Sub:M12", "Math.Neg:N1234"]}, {"position": 4, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:23", "Impact:A98", "Math.Sub:M0123", "Math.Neg:N12", "Attn:P2=99"]}, {"position": 4, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:0", "Impact:A9", "Math.Neg:N3", "Attn:P1=50", "Attn:P2=35", "Attn:P0=7", "Attn:P4=6"]}, {"position": 4, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:2", "Impact:A98", "Math.Sub:M13", "Math.Neg:N123", "Attn:P3=72", "Attn:P2=18", "Attn:P4=5", "Attn:P1=4"]}, {"position": 4, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:10", "Impact:A981", "Math.Sub:M123", "Math.Neg:N123"]}, {"position": 5, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:23", "Impact:A8761", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P4=99"]}, {"position": 5, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:1", "Impact:A71", "Math.Sub:M12", "Math.Neg:N134", "Attn:P3=95", "Attn:P4=3"]}, {"position": 5, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:1", "Impact:A764", "Math.Neg:N234", "Attn:P5=53", "Attn:P4=31", "Attn:P3=9", "Attn:P2=3"]}, {"position": 5, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:6", "Impact:A76", "Math.Sub:M12", "Math.Neg:N1234"]}, {"position": 6, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:5", "Impact:A6541", "Math.Sub:M123", "Math.Neg:N123", "Attn:P6=99"]}, {"position": 6, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:0", "Impact:A8765", "Math.Neg:N2", "Attn:P5=82", "Attn:P4=14", "Attn:P3=1"]}, {"position": 6, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:4", "Impact:A76541", "Math.Sub:M12", "Math.Neg:N12", "Attn:P6=88", "Attn:P5=6", "Attn:P4=2", "Attn:P3=2"]}, {"position": 6, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:3", "Impact:A7654", "Math.Sub:M3", "Math.Neg:N12"]}, {"position": 7, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:21", "Impact:A87651", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P5=99"]}, {"position": 7, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:3", "Impact:A6541", "Math.Sub:M123", "Math.Neg:N123", "Attn:P6=51", "Attn:P5=24", "Attn:P7=13", "Attn:P4=6"]}, {"position": 7, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:8", "Impact:A876541", "Math.Sub:M123", "Math.Neg:N1234"]}, {"position": 8, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:3", "Impact:A65431", "Math.Sub:M123", "Math.Neg:N12", "Attn:P7=83", "Attn:P6=13"]}, {"position": 8, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:0", "Impact:A4", "Math.Neg:N12", "Attn:P6=98"]}, {"position": 8, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:2", "Impact:A541", "Math.Sub:M12", "Math.Neg:N12", "Attn:P6=38", "Attn:P7=38", "Attn:P8=16", "Attn:P5=3"]}, {"position": 8, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:3", "Impact:A6543", "Math.Sub:M13", "Math.Neg:N12"]}, {"position": 9, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:0", "Impact:A86", "Math.Neg:N2", "Attn:P2=16", "Attn:P0=16", "Attn:P1=16", "Attn:P3=11"]}, {"position": 11, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:3", "Impact:A8764321", "Math.Sub:M12", "Math.Neg:N12", "Attn:P8=88", "Attn:P10=6", "Attn:P7=1", "Attn:P6=1"]}, {"position": 11, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:0", "Impact:A64", "Math.Neg:N2", "Attn:P8=32", "Attn:P9=27", "Attn:P7=18", "Attn:P10=8"]}, {"position": 11, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:2", "Impact:A432", "Math.Sub:M2", "Math.Neg:N12"]}, {"position": 12, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:21", "Impact:A865421", "Math.Sub:M123", "Math.Neg:N123", "Attn:P9=94", "Attn:P8=1"]}, {"position": 12, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:0", "Impact:A2", "Math.Sub:M2", "Attn:P8=97"]}, {"position": 12, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:0", "Impact:A41", "Math.Sub:M12", "Math.Neg:N2", "Attn:P10=38", "Attn:P9=32", "Attn:P8=11", "Attn:P0=5"]}, {"position": 12, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:9", "Impact:A321", "Math.Sub:M12", "Math.Neg:N123"]}, {"position": 13, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:11", "Impact:A43", "Math.Sub:M12", "Math.Neg:N12", "Attn:P7=98"]}, {"position": 13, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:0", "Impact:A3", "Math.Neg:N12", "Attn:P8=55", "Attn:P9=14", "Attn:P7=12", "Attn:P10=7"]}, {"position": 13, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:4", "Impact:A43", "Math.Sub:M12", "Math.Neg:N12"]}, {"position": 14, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:11", "Impact:A43", "Math.Sub:M12", "Math.Neg:N12", "Attn:P7=98"]}, {"position": 14, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:0", "Impact:A3", "Math.Neg:N12", "Attn:P8=56", "Attn:P9=14", "Attn:P7=13", "Attn:P10=6"]}, {"position": 14, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:4", "Impact:A43", "Math.Sub:M12", "Math.Neg:N12"]}, {"position": 16, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:12", "Impact:A2", "Math.Sub:M0123", "Math.Neg:N12", "Attn:P8=96"]}, {"position": 16, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:0", "Impact:A2", "Math.Sub:M1", "Attn:P9=63", "Attn:P10=17", "Attn:P8=10", "Attn:P7=2"]}, {"position": 16, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:6", "Impact:A2", "Math.Sub:M012", "Math.Neg:N12"]}, {"position": 18, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:12", "Impact:A21", "Math.Sub:M0123", "Math.Neg:N12", "Attn:P8=95"]}, {"position": 18, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:0", "Impact:A2", "Math.Sub:M1", "Attn:P9=57", "Attn:P10=20", "Attn:P8=8", "Attn:P18=2"]}, {"position": 18, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:6", "Impact:A2", "Math.Sub:M012", "Math.Neg:N12"]}, {"position": 21, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:7", "Impact:A91", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P12=98"]}, {"position": 21, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:44", "Impact:A91", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P0=52", "Attn:P11=48"]}, {"position": 21, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:8", "Impact:A91", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P12=59", "Attn:P13=31", "Attn:P14=6", "Attn:P1=1"]}, {"position": 21, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:48", "Impact:A9876543210", "Math.Sub:M0123", "Math.Neg:N123"]}, {"position": 21, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:2", "Impact:A1", "Math.Neg:N12", "Attn:P21=53", "Attn:P1=31", "Attn:P2=10", "Attn:P4=2"]}, {"position": 21, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:12", "Impact:A1", "Math.Sub:M0123", "Attn:P1=37", "Attn:P21=37", "Attn:P2=16", "Attn:P4=3"]}, {"position": 21, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:31", "Impact:A1", "Math.Sub:M123", "Math.Neg:N123"]}, {"position": 22, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:50", "Impact:A9876543210", "Math.Sub:M123", "Math.Neg:N1234", "Attn:P22=91", "Attn:P5=2", "Attn:P2=1", "Attn:P21=1"]}, {"position": 22, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:1", "Impact:A87654", "Math.Neg:N2", "Attn:P4=47", "Attn:P15=42", "Attn:P21=5", "Attn:P16=2"]}, {"position": 22, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:55", "Impact:A9876543210", "Math.Sub:M123", "Math.Neg:N1234"]}, {"position": 23, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:87", "Impact:A9", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P11=55", "Attn:P0=36", "Attn:P21=1", "Attn:P12=1"]}, {"position": 23, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:43", "Impact:A9", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P0=72", "Attn:P11=22"]}, {"position": 23, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:8", "Impact:A9", "Math.Sub:M123", "Math.Neg:N123", "Attn:P12=52", "Attn:P13=31", "Attn:P14=8", "Attn:P1=2"]}, {"position": 23, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:80", "Impact:A9", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 23, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:33", "Impact:A9", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P22=31", "Attn:P21=13", "Attn:P3=9", "Attn:P1=9"]}, {"position": 23, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:24", "Impact:A9", "Math.Sub:M0123", "Math.Neg:N12", "Attn:P21=53", "Attn:P1=34", "Attn:P2=9", "Attn:P4=3"]}, {"position": 23, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:12", "Impact:A9", "Math.Sub:M12", "Math.Neg:N123", "Attn:P21=43", "Attn:P1=39", "Attn:P2=10", "Attn:P4=4"]}, {"position": 23, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:73", "Impact:A9", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 24, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:90", "Impact:A8", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P12=42", "Attn:P1=25", "Attn:P21=12", "Attn:P13=2"]}, {"position": 24, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:49", "Impact:A8", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P1=66", "Attn:P22=6", "Attn:P10=6", "Attn:P12=4"]}, {"position": 24, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:30", "Impact:A8", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P13=63", "Attn:P14=27", "Attn:P12=2", "Attn:P15=2"]}, {"position": 24, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:84", "Impact:A8", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 24, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:42", "Impact:A8", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P22=35", "Attn:P21=11", "Attn:P1=9", "Attn:P0=7"]}, {"position": 24, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:31", "Impact:A8", "Math.Sub:M0123", "Math.Neg:N12", "Attn:P22=50", "Attn:P4=45", "Attn:P3=3"]}, {"position": 24, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:19", "Impact:A8", "Math.Sub:M123", "Math.Neg:N123", "Attn:P4=48", "Attn:P22=38", "Attn:P3=8", "Attn:P21=1"]}, {"position": 24, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:77", "Impact:A8", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 25, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:89", "Impact:A7", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P13=39", "Attn:P2=24", "Attn:P10=12", "Attn:P22=5"]}, {"position": 25, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:57", "Impact:A7", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P2=75", "Attn:P22=7", "Attn:P13=2", "Attn:P1=2"]}, {"position": 25, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:32", "Impact:A7", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P14=62", "Attn:P15=22", "Attn:P16=7", "Attn:P13=2"]}, {"position": 25, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:86", "Impact:A7", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 25, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:47", "Impact:A7", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=32", "Attn:P21=12", "Attn:P0=9", "Attn:P3=8"]}, {"position": 25, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:35", "Impact:A7", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=52", "Attn:P3=40", "Attn:P5=8"]}, {"position": 25, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:19", "Impact:A7", "Math.Sub:M23", "Math.Neg:N1234", "Attn:P3=50", "Attn:P22=37", "Attn:P5=9", "Attn:P7=1"]}, {"position": 25, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:78", "Impact:A7", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 26, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:89", "Impact:A6", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P14=36", "Attn:P3=24", "Attn:P10=15", "Attn:P22=3"]}, {"position": 26, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:52", "Impact:A6", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P3=69", "Attn:P10=11", "Attn:P22=6", "Attn:P15=1"]}, {"position": 26, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:21", "Impact:A6", "Math.Sub:M012", "Math.Neg:N123", "Attn:P15=60", "Attn:P16=26", "Attn:P17=6", "Attn:P14=2"]}, {"position": 26, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:87", "Impact:A6", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 26, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:47", "Impact:A6", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=35", "Attn:P21=13", "Attn:P1=9", "Attn:P0=7"]}, {"position": 26, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:32", "Impact:A6", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=51", "Attn:P5=44", "Attn:P7=4"]}, {"position": 26, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:18", "Impact:A6", "Math.Sub:M12", "Math.Neg:N1234", "Attn:P5=51", "Attn:P22=39", "Attn:P7=6", "Attn:P6=1"]}, {"position": 26, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:67", "Impact:A6", "Math.Sub:M0123", "Math.Neg:N123"]}, {"position": 27, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:88", "Impact:A5", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P15=38", "Attn:P4=21", "Attn:P10=14", "Attn:P23=3"]}, {"position": 27, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:60", "Impact:A5", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P4=75", "Attn:P15=8", "Attn:P10=6", "Attn:P22=3"]}, {"position": 27, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:29", "Impact:A5", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P16=62", "Attn:P17=25", "Attn:P18=5", "Attn:P15=2"]}, {"position": 27, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:89", "Impact:A5", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 27, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:47", "Impact:A5", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=34", "Attn:P21=13", "Attn:P1=10", "Attn:P0=8"]}, {"position": 27, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:33", "Impact:A5", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=49", "Attn:P7=46", "Attn:P6=4"]}, {"position": 27, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:23", "Impact:A5", "Math.Sub:M123", "Math.Neg:N1234", "Attn:P7=52", "Attn:P22=40", "Attn:P6=5"]}, {"position": 27, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:69", "Impact:A5", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 28, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:89", "Impact:A4", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P16=38", "Attn:P5=23", "Attn:P10=14", "Attn:P22=3"]}, {"position": 28, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:52", "Impact:A4", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P5=67", "Attn:P22=10", "Attn:P10=7", "Attn:P4=2"]}, {"position": 28, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:27", "Impact:A4", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P17=61", "Attn:P18=25", "Attn:P19=5", "Attn:P16=2"]}, {"position": 28, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:88", "Impact:A4", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 28, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:43", "Impact:A4", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=35", "Attn:P21=11", "Attn:P1=9", "Attn:P0=7"]}, {"position": 28, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:32", "Impact:A4", "Math.Sub:M0123", "Math.Neg:N12", "Attn:P22=55", "Attn:P6=37", "Attn:P8=7"]}, {"position": 28, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:19", "Impact:A4", "Math.Sub:M123", "Math.Neg:N1234", "Attn:P6=44", "Attn:P22=42", "Attn:P8=10"]}, {"position": 28, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:64", "Impact:A4", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 29, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:87", "Impact:A3", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P17=36", "Attn:P6=23", "Attn:P10=17", "Attn:P22=2"]}, {"position": 29, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:46", "Impact:A3", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P6=54", "Attn:P10=25", "Attn:P22=5", "Attn:P18=2"]}, {"position": 29, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:27", "Impact:A3", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P18=58", "Attn:P19=26", "Attn:P20=7", "Attn:P17=2"]}, {"position": 29, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:86", "Impact:A3", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 29, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:54", "Impact:A3", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P22=36", "Attn:P21=13", "Attn:P1=10", "Attn:P0=7"]}, {"position": 29, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:34", "Impact:A3", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=51", "Attn:P14=21", "Attn:P13=20", "Attn:P11=5"]}, {"position": 29, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:19", "Impact:A3", "Math.Sub:M123", "Math.Neg:N1234", "Attn:P22=38", "Attn:P14=26", "Attn:P13=24", "Attn:P11=6"]}, {"position": 29, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:65", "Impact:A3", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 30, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:89", "Impact:A2", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P18=45", "Attn:P7=29", "Attn:P10=3", "Attn:P19=3"]}, {"position": 30, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:43", "Impact:A2", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P7=67", "Attn:P22=13", "Attn:P10=11"]}, {"position": 30, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:23", "Impact:A2", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P19=58", "Attn:P20=31", "Attn:P18=2"]}, {"position": 30, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:87", "Impact:A2", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 30, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:48", "Impact:A2", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=35", "Attn:P21=11", "Attn:P1=9", "Attn:P0=8"]}, {"position": 30, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:34", "Impact:A2", "Math.Sub:M012", "Math.Neg:N123", "Attn:P22=46", "Attn:P16=22", "Attn:P18=21", "Attn:P12=5"]}, {"position": 30, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:25", "Impact:A2", "Math.Sub:M12", "Math.Neg:N1234", "Attn:P22=36", "Attn:P16=25", "Attn:P18=23", "Attn:P12=8"]}, {"position": 30, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:65", "Impact:A2", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 31, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:90", "Impact:A1", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P19=48", "Attn:P8=29", "Attn:P20=3", "Attn:P10=2"]}, {"position": 31, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:48", "Impact:A1", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P8=87", "Attn:P22=5", "Attn:P9=1"]}, {"position": 31, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:27", "Impact:A1", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P20=46", "Attn:P22=18", "Attn:P19=4", "Attn:P10=3"]}, {"position": 31, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:88", "Impact:A1", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 31, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:43", "Impact:A1", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=36", "Attn:P21=11", "Attn:P1=10", "Attn:P3=7"]}, {"position": 31, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:44", "Impact:A1", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=52", "Attn:P12=41", "Attn:P31=1"]}, {"position": 31, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:27", "Impact:A1", "Math.Sub:M12", "Math.Neg:N1234", "Attn:P12=54", "Attn:P22=37", "Attn:P31=1"]}, {"position": 31, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:68", "Impact:A1", "Math.Sub:M0123", "Math.Neg:N1234"]}, {"position": 32, "layer": 0, "is_head": true, "num": 0, "tags": ["Fail%:92", "Impact:A0", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P20=54", "Attn:P9=27", "Attn:P10=3", "Attn:P23=2"]}, {"position": 32, "layer": 0, "is_head": true, "num": 1, "tags": ["Fail%:49", "Impact:A0", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P9=90", "Attn:P8=2"]}, {"position": 32, "layer": 0, "is_head": true, "num": 2, "tags": ["Fail%:18", "Impact:A0", "Math.Sub:M0123", "Math.Neg:N123", "Attn:P22=28", "Attn:P23=15", "Attn:P9=13", "Attn:P20=8"]}, {"position": 32, "layer": 0, "is_head": false, "num": 0, "tags": ["Fail%:85", "Impact:A0", "Math.Sub:M0123", "Math.Neg:N123"]}, {"position": 32, "layer": 1, "is_head": true, "num": 0, "tags": ["Fail%:45", "Impact:A0", "Math.Sub:M0123", "Math.Neg:N1234", "Attn:P22=39", "Attn:P21=11", "Attn:P1=8", "Attn:P0=7"]}, {"position": 32, "layer": 1, "is_head": true, "num": 1, "tags": ["Fail%:45", "Impact:A0", "Math.Sub:M0123", "Attn:P22=60", "Attn:P12=10", "Attn:P32=8", "Attn:P21=2"]}, {"position": 32, "layer": 1, "is_head": true, "num": 2, "tags": ["Fail%:48", "Impact:A0", "Math.Neg:N1234", "Attn:P22=39", "Attn:P12=29", "Attn:P32=11", "Attn:P7=2"]}, {"position": 32, "layer": 1, "is_head": false, "num": 0, "tags": ["Fail%:61", "Impact:A0", "Math.Sub:M0123", "Math.Neg:N1234"]}]
features.json ADDED
@@ -0,0 +1 @@
 
 
1
+ [{"position": 1, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 1, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 1, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 2, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 2, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 2, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 3, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 3, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 3, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 4, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 4, "layer": 0, "is_head": true, "num": 1, "tags": []}, {"position": 4, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 4, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 5, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 5, "layer": 0, "is_head": true, "num": 1, "tags": []}, {"position": 5, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 5, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 6, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 6, "layer": 0, "is_head": true, "num": 1, "tags": []}, {"position": 6, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 6, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 7, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 7, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 7, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 8, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 8, "layer": 0, "is_head": true, "num": 1, "tags": []}, {"position": 8, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 8, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 9, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 11, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 11, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 11, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 12, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 12, "layer": 0, "is_head": true, "num": 1, "tags": []}, {"position": 12, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 12, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 13, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 13, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 13, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 14, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 14, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 14, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 16, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 16, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 16, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 18, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 18, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 18, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 21, "layer": 0, "is_head": true, "num": 0, "tags": []}, {"position": 21, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:A9.MT", "Algo:A9.GT"]}, {"position": 21, "layer": 0, "is_head": true, "num": 2, "tags": ["Algo:A8.MT"]}, {"position": 21, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 21, "layer": 1, "is_head": true, "num": 1, "tags": []}, {"position": 21, "layer": 1, "is_head": true, "num": 2, "tags": []}, {"position": 21, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 22, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 22, "layer": 0, "is_head": true, "num": 1, "tags": []}, {"position": 22, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 23, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A9.MD", "Algo:A9.ND.A9"]}, {"position": 23, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:A9.MD"]}, {"position": 23, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 23, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 23, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 23, "layer": 1, "is_head": true, "num": 1, "tags": []}, {"position": 23, "layer": 1, "is_head": true, "num": 2, "tags": []}, {"position": 23, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 24, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A8.MD", "Algo:A8.ND"]}, {"position": 24, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:A8.MD", "Algo:SGN", "Algo:A8.ND"]}, {"position": 24, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 24, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 24, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 24, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 24, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 24, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 25, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A7.MD", "Algo:SGN", "Algo:A7.ND"]}, {"position": 25, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:A7.MD", "Algo:SGN", "Algo:A7.ND"]}, {"position": 25, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 25, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 25, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 25, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 25, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 25, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 26, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A6.MD.A6", "Algo:SGN", "Algo:A6.ND.A6"]}, {"position": 26, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 26, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 26, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 26, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 26, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 26, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 26, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 27, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A5.MD", "Algo:A5.ND"]}, {"position": 27, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:A5.MD", "Algo:SGN", "Algo:A5.ND"]}, {"position": 27, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 27, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 27, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 27, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 27, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 27, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 28, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A4.MD.A4", "Algo:SGN", "Algo:A4.ND.A4"]}, {"position": 28, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 28, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 28, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 28, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 28, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 28, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 28, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 29, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A3.MD.A3", "Algo:SGN", "Algo:A3.ND.A3"]}, {"position": 29, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 29, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 29, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 29, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 29, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 29, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 29, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 30, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A2.MD.A2", "Algo:A2.ND.A2"]}, {"position": 30, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 30, "layer": 0, "is_head": true, "num": 2, "tags": []}, {"position": 30, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 30, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 30, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 30, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 30, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 31, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A1.MD.A1", "Algo:A1.ND.A1"]}, {"position": 31, "layer": 0, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 31, "layer": 0, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 31, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 31, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 31, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 31, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 31, "layer": 1, "is_head": false, "num": 0, "tags": []}, {"position": 32, "layer": 0, "is_head": true, "num": 0, "tags": ["Algo:A0.MD.A0", "Algo:A0.ND.A0"]}, {"position": 32, "layer": 0, "is_head": true, "num": 1, "tags": []}, {"position": 32, "layer": 0, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 32, "layer": 0, "is_head": false, "num": 0, "tags": []}, {"position": 32, "layer": 1, "is_head": true, "num": 0, "tags": ["Algo:SGN"]}, {"position": 32, "layer": 1, "is_head": true, "num": 1, "tags": ["Algo:SGN"]}, {"position": 32, "layer": 1, "is_head": true, "num": 2, "tags": ["Algo:SGN"]}, {"position": 32, "layer": 1, "is_head": false, "num": 0, "tags": []}]
model.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd9d5d8a9ccf0fe9d6acade00c5cb2bbec5118b67b7dec25720946411de52c9c
3
+ size 25176450
training_loss.json ADDED
The diff for this file is too large to render. See raw diff