thkim0305 commited on
Commit
b366ef8
·
verified ·
1 Parent(s): 30d9189

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round10.pth +3 -0
  2. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round12.pth +3 -0
  3. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round15.pth +3 -0
  4. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round17.pth +3 -0
  5. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round2.pth +3 -0
  6. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round20.pth +3 -0
  7. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round5.pth +3 -0
  8. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round7.pth +3 -0
  9. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_trainer_state.json +392 -0
  10. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round10.pth +3 -0
  11. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round12.pth +3 -0
  12. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round15.pth +3 -0
  13. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round17.pth +3 -0
  14. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round2.pth +3 -0
  15. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round20.pth +3 -0
  16. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round5.pth +3 -0
  17. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round7.pth +3 -0
  18. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_trainer_state.json +392 -0
  19. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round10.pth +3 -0
  20. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round12.pth +3 -0
  21. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round15.pth +3 -0
  22. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round17.pth +3 -0
  23. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round2.pth +3 -0
  24. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round20.pth +3 -0
  25. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round5.pth +3 -0
  26. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round7.pth +3 -0
  27. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_trainer_state.json +392 -0
  28. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round10.pth +3 -0
  29. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round12.pth +3 -0
  30. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round15.pth +3 -0
  31. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round17.pth +3 -0
  32. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round2.pth +3 -0
  33. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round20.pth +3 -0
  34. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round5.pth +3 -0
  35. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round7.pth +3 -0
  36. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_trainer_state.json +392 -0
  37. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round10.pth +3 -0
  38. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round12.pth +3 -0
  39. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round15.pth +3 -0
  40. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round17.pth +3 -0
  41. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round2.pth +3 -0
  42. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round20.pth +3 -0
  43. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round5.pth +3 -0
  44. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round7.pth +3 -0
  45. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_trainer_state.json +392 -0
  46. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round10.pth +3 -0
  47. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round12.pth +3 -0
  48. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round15.pth +3 -0
  49. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round17.pth +3 -0
  50. client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round2.pth +3 -0
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round10.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7281b71feb874201c3510fd2532c735c41834e0ea57f60b3a19d83a172c8062d
3
+ size 184221358
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round12.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43f3b3c81a25994ed2450656aad07f50b0d6a686213e90ee5936f5227e75b924
3
+ size 184221358
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round15.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02634f446483c9401811d202c4fb26c234e200c1b5bd389dca72585c2d61d034
3
+ size 184221358
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round17.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b593199404c65a7fb376d970ec893e60e94daef7a5ae8eed3ca517e9275fe0c
3
+ size 184221358
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f312bff96288926254f6abc38cb4e1313e8d188484ebbbeadf5f59357ce0f3e3
3
+ size 184220842
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round20.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03270abac1135f1e20d23bcbaf393ffd4157017414b071ece5ec558a5319692a
3
+ size 184221358
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc2eaa3387065398c5c4e6cb687acf55c8c80d95eddaf5452a6456e4dd5b1d6f
3
+ size 184220842
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_client_model_round7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bea9fbbcea6ab83084b33a90fc70ebbcc775b9472862c5f46300a085976a036
3
+ size 184220842
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/0_trainer_state.json ADDED
@@ -0,0 +1,392 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 100,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02,
13
+ "grad_norm": 4.327882289886475,
14
+ "learning_rate": 2e-05,
15
+ "loss": 0.3049,
16
+ "step": 2
17
+ },
18
+ {
19
+ "epoch": 0.04,
20
+ "grad_norm": 4.293352127075195,
21
+ "learning_rate": 2e-05,
22
+ "loss": 0.4343,
23
+ "step": 4
24
+ },
25
+ {
26
+ "epoch": 0.06,
27
+ "grad_norm": 8.278122901916504,
28
+ "learning_rate": 2e-05,
29
+ "loss": 0.464,
30
+ "step": 6
31
+ },
32
+ {
33
+ "epoch": 0.08,
34
+ "grad_norm": 5.572937488555908,
35
+ "learning_rate": 2e-05,
36
+ "loss": 1.1184,
37
+ "step": 8
38
+ },
39
+ {
40
+ "epoch": 0.1,
41
+ "grad_norm": 1.2564692497253418,
42
+ "learning_rate": 2e-05,
43
+ "loss": 0.1493,
44
+ "step": 10
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "grad_norm": 2.756223201751709,
49
+ "learning_rate": 2e-05,
50
+ "loss": 0.1429,
51
+ "step": 12
52
+ },
53
+ {
54
+ "epoch": 0.14,
55
+ "grad_norm": 6.3286638259887695,
56
+ "learning_rate": 2e-05,
57
+ "loss": 1.0986,
58
+ "step": 14
59
+ },
60
+ {
61
+ "epoch": 0.16,
62
+ "grad_norm": 3.1671125888824463,
63
+ "learning_rate": 2e-05,
64
+ "loss": 0.421,
65
+ "step": 16
66
+ },
67
+ {
68
+ "epoch": 0.18,
69
+ "grad_norm": 12.160600662231445,
70
+ "learning_rate": 2e-05,
71
+ "loss": 0.6418,
72
+ "step": 18
73
+ },
74
+ {
75
+ "epoch": 0.2,
76
+ "grad_norm": 2.2388546466827393,
77
+ "learning_rate": 2e-05,
78
+ "loss": 0.3821,
79
+ "step": 20
80
+ },
81
+ {
82
+ "epoch": 0.22,
83
+ "grad_norm": 4.057492733001709,
84
+ "learning_rate": 2e-05,
85
+ "loss": 0.2804,
86
+ "step": 22
87
+ },
88
+ {
89
+ "epoch": 0.24,
90
+ "grad_norm": 2.4171645641326904,
91
+ "learning_rate": 2e-05,
92
+ "loss": 0.2768,
93
+ "step": 24
94
+ },
95
+ {
96
+ "epoch": 0.26,
97
+ "grad_norm": 4.334846019744873,
98
+ "learning_rate": 2e-05,
99
+ "loss": 0.5183,
100
+ "step": 26
101
+ },
102
+ {
103
+ "epoch": 0.28,
104
+ "grad_norm": 5.31112003326416,
105
+ "learning_rate": 2e-05,
106
+ "loss": 0.3826,
107
+ "step": 28
108
+ },
109
+ {
110
+ "epoch": 0.3,
111
+ "grad_norm": 8.947898864746094,
112
+ "learning_rate": 2e-05,
113
+ "loss": 0.7833,
114
+ "step": 30
115
+ },
116
+ {
117
+ "epoch": 0.32,
118
+ "grad_norm": 4.790311336517334,
119
+ "learning_rate": 2e-05,
120
+ "loss": 0.7227,
121
+ "step": 32
122
+ },
123
+ {
124
+ "epoch": 0.34,
125
+ "grad_norm": 1.9647458791732788,
126
+ "learning_rate": 2e-05,
127
+ "loss": 0.2151,
128
+ "step": 34
129
+ },
130
+ {
131
+ "epoch": 0.36,
132
+ "grad_norm": 7.152100086212158,
133
+ "learning_rate": 2e-05,
134
+ "loss": 0.6738,
135
+ "step": 36
136
+ },
137
+ {
138
+ "epoch": 0.38,
139
+ "grad_norm": 9.002793312072754,
140
+ "learning_rate": 2e-05,
141
+ "loss": 1.2422,
142
+ "step": 38
143
+ },
144
+ {
145
+ "epoch": 0.4,
146
+ "grad_norm": 4.538879871368408,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.7256,
149
+ "step": 40
150
+ },
151
+ {
152
+ "epoch": 0.42,
153
+ "grad_norm": 7.2817158699035645,
154
+ "learning_rate": 2e-05,
155
+ "loss": 0.4922,
156
+ "step": 42
157
+ },
158
+ {
159
+ "epoch": 0.44,
160
+ "grad_norm": 8.116897583007812,
161
+ "learning_rate": 2e-05,
162
+ "loss": 0.2341,
163
+ "step": 44
164
+ },
165
+ {
166
+ "epoch": 0.46,
167
+ "grad_norm": 5.7875776290893555,
168
+ "learning_rate": 2e-05,
169
+ "loss": 0.5828,
170
+ "step": 46
171
+ },
172
+ {
173
+ "epoch": 0.48,
174
+ "grad_norm": 2.9006869792938232,
175
+ "learning_rate": 2e-05,
176
+ "loss": 0.2646,
177
+ "step": 48
178
+ },
179
+ {
180
+ "epoch": 0.5,
181
+ "grad_norm": 5.325571060180664,
182
+ "learning_rate": 2e-05,
183
+ "loss": 0.5073,
184
+ "step": 50
185
+ },
186
+ {
187
+ "epoch": 0.52,
188
+ "grad_norm": 1.7935487031936646,
189
+ "learning_rate": 2e-05,
190
+ "loss": 0.3458,
191
+ "step": 52
192
+ },
193
+ {
194
+ "epoch": 0.54,
195
+ "grad_norm": 6.689419269561768,
196
+ "learning_rate": 2e-05,
197
+ "loss": 0.9642,
198
+ "step": 54
199
+ },
200
+ {
201
+ "epoch": 0.56,
202
+ "grad_norm": 9.98328685760498,
203
+ "learning_rate": 2e-05,
204
+ "loss": 0.7798,
205
+ "step": 56
206
+ },
207
+ {
208
+ "epoch": 0.58,
209
+ "grad_norm": 10.168957710266113,
210
+ "learning_rate": 2e-05,
211
+ "loss": 0.3466,
212
+ "step": 58
213
+ },
214
+ {
215
+ "epoch": 0.6,
216
+ "grad_norm": 5.621067523956299,
217
+ "learning_rate": 2e-05,
218
+ "loss": 0.4138,
219
+ "step": 60
220
+ },
221
+ {
222
+ "epoch": 0.62,
223
+ "grad_norm": 16.461015701293945,
224
+ "learning_rate": 2e-05,
225
+ "loss": 0.963,
226
+ "step": 62
227
+ },
228
+ {
229
+ "epoch": 0.64,
230
+ "grad_norm": 3.5031635761260986,
231
+ "learning_rate": 2e-05,
232
+ "loss": 0.6482,
233
+ "step": 64
234
+ },
235
+ {
236
+ "epoch": 0.66,
237
+ "grad_norm": 1.3376978635787964,
238
+ "learning_rate": 2e-05,
239
+ "loss": 0.1724,
240
+ "step": 66
241
+ },
242
+ {
243
+ "epoch": 0.68,
244
+ "grad_norm": 1.3010202646255493,
245
+ "learning_rate": 2e-05,
246
+ "loss": 0.2854,
247
+ "step": 68
248
+ },
249
+ {
250
+ "epoch": 0.7,
251
+ "grad_norm": 8.061728477478027,
252
+ "learning_rate": 2e-05,
253
+ "loss": 0.4303,
254
+ "step": 70
255
+ },
256
+ {
257
+ "epoch": 0.72,
258
+ "grad_norm": 5.248547077178955,
259
+ "learning_rate": 2e-05,
260
+ "loss": 0.5894,
261
+ "step": 72
262
+ },
263
+ {
264
+ "epoch": 0.74,
265
+ "grad_norm": 4.7633819580078125,
266
+ "learning_rate": 2e-05,
267
+ "loss": 0.6274,
268
+ "step": 74
269
+ },
270
+ {
271
+ "epoch": 0.76,
272
+ "grad_norm": 1.664157509803772,
273
+ "learning_rate": 2e-05,
274
+ "loss": 0.3461,
275
+ "step": 76
276
+ },
277
+ {
278
+ "epoch": 0.78,
279
+ "grad_norm": 2.7871382236480713,
280
+ "learning_rate": 2e-05,
281
+ "loss": 0.5779,
282
+ "step": 78
283
+ },
284
+ {
285
+ "epoch": 0.8,
286
+ "grad_norm": 5.3619914054870605,
287
+ "learning_rate": 2e-05,
288
+ "loss": 0.7158,
289
+ "step": 80
290
+ },
291
+ {
292
+ "epoch": 0.82,
293
+ "grad_norm": 4.908105373382568,
294
+ "learning_rate": 2e-05,
295
+ "loss": 0.5618,
296
+ "step": 82
297
+ },
298
+ {
299
+ "epoch": 0.84,
300
+ "grad_norm": 4.174102306365967,
301
+ "learning_rate": 2e-05,
302
+ "loss": 0.645,
303
+ "step": 84
304
+ },
305
+ {
306
+ "epoch": 0.86,
307
+ "grad_norm": 5.976959705352783,
308
+ "learning_rate": 2e-05,
309
+ "loss": 0.5615,
310
+ "step": 86
311
+ },
312
+ {
313
+ "epoch": 0.88,
314
+ "grad_norm": 5.8181586265563965,
315
+ "learning_rate": 2e-05,
316
+ "loss": 0.6136,
317
+ "step": 88
318
+ },
319
+ {
320
+ "epoch": 0.9,
321
+ "grad_norm": 2.994356155395508,
322
+ "learning_rate": 2e-05,
323
+ "loss": 0.196,
324
+ "step": 90
325
+ },
326
+ {
327
+ "epoch": 0.92,
328
+ "grad_norm": 4.384360313415527,
329
+ "learning_rate": 2e-05,
330
+ "loss": 0.2452,
331
+ "step": 92
332
+ },
333
+ {
334
+ "epoch": 0.94,
335
+ "grad_norm": 3.3636257648468018,
336
+ "learning_rate": 2e-05,
337
+ "loss": 0.3809,
338
+ "step": 94
339
+ },
340
+ {
341
+ "epoch": 0.96,
342
+ "grad_norm": 3.677785634994507,
343
+ "learning_rate": 2e-05,
344
+ "loss": 0.3801,
345
+ "step": 96
346
+ },
347
+ {
348
+ "epoch": 0.98,
349
+ "grad_norm": 6.116004467010498,
350
+ "learning_rate": 2e-05,
351
+ "loss": 0.5769,
352
+ "step": 98
353
+ },
354
+ {
355
+ "epoch": 1.0,
356
+ "grad_norm": 5.415121555328369,
357
+ "learning_rate": 2e-05,
358
+ "loss": 0.385,
359
+ "step": 100
360
+ },
361
+ {
362
+ "epoch": 1.0,
363
+ "step": 100,
364
+ "total_flos": 6589214774788096.0,
365
+ "train_loss": 0.5162237358093261,
366
+ "train_runtime": 121.9606,
367
+ "train_samples_per_second": 3.28,
368
+ "train_steps_per_second": 0.82
369
+ }
370
+ ],
371
+ "logging_steps": 2,
372
+ "max_steps": 100,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 1,
375
+ "save_steps": 500,
376
+ "stateful_callbacks": {
377
+ "TrainerControl": {
378
+ "args": {
379
+ "should_epoch_stop": false,
380
+ "should_evaluate": false,
381
+ "should_log": false,
382
+ "should_save": false,
383
+ "should_training_stop": false
384
+ },
385
+ "attributes": {}
386
+ }
387
+ },
388
+ "total_flos": 6589214774788096.0,
389
+ "train_batch_size": 2,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round10.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4b7524ff1cdb9c623c9f04b1b2f456ce956f8333f742e367f0627615c523df0
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round12.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b14de6a6ff83a367693fffd74a202f81b6610ae9866eca998fee857876b57cc6
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round15.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc604f9b5b83debe041f5baec0917bf5f25da3d2400cc3642da153b1a214c1a0
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round17.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fab435c732539a8ae5d19b6f27b57b14502392076a996005e26331e5c8488f24
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45faca8a1b9116373952a8ef00712fa1e379398914a28043649b825dc9345f5b
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round20.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e371888600f40c5bfd4398e7526aab31e42df408736003ec9c070aa0a7ed0d7e
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac5f9b83f02f09f963455cb0e4dbe43fe7930e2e9b1fd8d2976ffad70b1f71ab
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_client_model_round7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7728b781894e5139abed7ddf3350ce0c8d77c0f74d4a37fb6fa9db4f9b6b07b9
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/1_trainer_state.json ADDED
@@ -0,0 +1,392 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 100,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02,
13
+ "grad_norm": 8.241416931152344,
14
+ "learning_rate": 2e-05,
15
+ "loss": 0.8475,
16
+ "step": 2
17
+ },
18
+ {
19
+ "epoch": 0.04,
20
+ "grad_norm": 4.8631792068481445,
21
+ "learning_rate": 2e-05,
22
+ "loss": 0.6782,
23
+ "step": 4
24
+ },
25
+ {
26
+ "epoch": 0.06,
27
+ "grad_norm": 7.5998454093933105,
28
+ "learning_rate": 2e-05,
29
+ "loss": 0.5532,
30
+ "step": 6
31
+ },
32
+ {
33
+ "epoch": 0.08,
34
+ "grad_norm": 2.5101563930511475,
35
+ "learning_rate": 2e-05,
36
+ "loss": 0.5537,
37
+ "step": 8
38
+ },
39
+ {
40
+ "epoch": 0.1,
41
+ "grad_norm": 4.472541332244873,
42
+ "learning_rate": 2e-05,
43
+ "loss": 0.6426,
44
+ "step": 10
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "grad_norm": 4.083187103271484,
49
+ "learning_rate": 2e-05,
50
+ "loss": 0.5195,
51
+ "step": 12
52
+ },
53
+ {
54
+ "epoch": 0.14,
55
+ "grad_norm": 5.575020790100098,
56
+ "learning_rate": 2e-05,
57
+ "loss": 0.4822,
58
+ "step": 14
59
+ },
60
+ {
61
+ "epoch": 0.16,
62
+ "grad_norm": 4.933296203613281,
63
+ "learning_rate": 2e-05,
64
+ "loss": 0.5742,
65
+ "step": 16
66
+ },
67
+ {
68
+ "epoch": 0.18,
69
+ "grad_norm": 2.017449378967285,
70
+ "learning_rate": 2e-05,
71
+ "loss": 0.4663,
72
+ "step": 18
73
+ },
74
+ {
75
+ "epoch": 0.2,
76
+ "grad_norm": 6.998116493225098,
77
+ "learning_rate": 2e-05,
78
+ "loss": 0.582,
79
+ "step": 20
80
+ },
81
+ {
82
+ "epoch": 0.22,
83
+ "grad_norm": 2.1770007610321045,
84
+ "learning_rate": 2e-05,
85
+ "loss": 0.5249,
86
+ "step": 22
87
+ },
88
+ {
89
+ "epoch": 0.24,
90
+ "grad_norm": 5.969364166259766,
91
+ "learning_rate": 2e-05,
92
+ "loss": 0.623,
93
+ "step": 24
94
+ },
95
+ {
96
+ "epoch": 0.26,
97
+ "grad_norm": 1.237125277519226,
98
+ "learning_rate": 2e-05,
99
+ "loss": 0.3623,
100
+ "step": 26
101
+ },
102
+ {
103
+ "epoch": 0.28,
104
+ "grad_norm": 6.773047924041748,
105
+ "learning_rate": 2e-05,
106
+ "loss": 0.4094,
107
+ "step": 28
108
+ },
109
+ {
110
+ "epoch": 0.3,
111
+ "grad_norm": 2.925546169281006,
112
+ "learning_rate": 2e-05,
113
+ "loss": 0.6287,
114
+ "step": 30
115
+ },
116
+ {
117
+ "epoch": 0.32,
118
+ "grad_norm": 3.002979278564453,
119
+ "learning_rate": 2e-05,
120
+ "loss": 0.7397,
121
+ "step": 32
122
+ },
123
+ {
124
+ "epoch": 0.34,
125
+ "grad_norm": 1.8541330099105835,
126
+ "learning_rate": 2e-05,
127
+ "loss": 0.5017,
128
+ "step": 34
129
+ },
130
+ {
131
+ "epoch": 0.36,
132
+ "grad_norm": 7.924871921539307,
133
+ "learning_rate": 2e-05,
134
+ "loss": 0.6328,
135
+ "step": 36
136
+ },
137
+ {
138
+ "epoch": 0.38,
139
+ "grad_norm": 2.390252113342285,
140
+ "learning_rate": 2e-05,
141
+ "loss": 0.3247,
142
+ "step": 38
143
+ },
144
+ {
145
+ "epoch": 0.4,
146
+ "grad_norm": 3.807579278945923,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.4795,
149
+ "step": 40
150
+ },
151
+ {
152
+ "epoch": 0.42,
153
+ "grad_norm": 3.4766390323638916,
154
+ "learning_rate": 2e-05,
155
+ "loss": 0.4468,
156
+ "step": 42
157
+ },
158
+ {
159
+ "epoch": 0.44,
160
+ "grad_norm": 1.4840394258499146,
161
+ "learning_rate": 2e-05,
162
+ "loss": 0.5312,
163
+ "step": 44
164
+ },
165
+ {
166
+ "epoch": 0.46,
167
+ "grad_norm": 3.136503219604492,
168
+ "learning_rate": 2e-05,
169
+ "loss": 0.4313,
170
+ "step": 46
171
+ },
172
+ {
173
+ "epoch": 0.48,
174
+ "grad_norm": 2.3651793003082275,
175
+ "learning_rate": 2e-05,
176
+ "loss": 0.3152,
177
+ "step": 48
178
+ },
179
+ {
180
+ "epoch": 0.5,
181
+ "grad_norm": 3.1154613494873047,
182
+ "learning_rate": 2e-05,
183
+ "loss": 0.6135,
184
+ "step": 50
185
+ },
186
+ {
187
+ "epoch": 0.52,
188
+ "grad_norm": 5.85906982421875,
189
+ "learning_rate": 2e-05,
190
+ "loss": 0.9023,
191
+ "step": 52
192
+ },
193
+ {
194
+ "epoch": 0.54,
195
+ "grad_norm": 5.495807647705078,
196
+ "learning_rate": 2e-05,
197
+ "loss": 0.5264,
198
+ "step": 54
199
+ },
200
+ {
201
+ "epoch": 0.56,
202
+ "grad_norm": 2.5824015140533447,
203
+ "learning_rate": 2e-05,
204
+ "loss": 0.2805,
205
+ "step": 56
206
+ },
207
+ {
208
+ "epoch": 0.58,
209
+ "grad_norm": 6.252473831176758,
210
+ "learning_rate": 2e-05,
211
+ "loss": 0.4327,
212
+ "step": 58
213
+ },
214
+ {
215
+ "epoch": 0.6,
216
+ "grad_norm": 5.942077159881592,
217
+ "learning_rate": 2e-05,
218
+ "loss": 0.4004,
219
+ "step": 60
220
+ },
221
+ {
222
+ "epoch": 0.62,
223
+ "grad_norm": 5.180813789367676,
224
+ "learning_rate": 2e-05,
225
+ "loss": 0.864,
226
+ "step": 62
227
+ },
228
+ {
229
+ "epoch": 0.64,
230
+ "grad_norm": 3.008401870727539,
231
+ "learning_rate": 2e-05,
232
+ "loss": 0.292,
233
+ "step": 64
234
+ },
235
+ {
236
+ "epoch": 0.66,
237
+ "grad_norm": 1.5255389213562012,
238
+ "learning_rate": 2e-05,
239
+ "loss": 0.7554,
240
+ "step": 66
241
+ },
242
+ {
243
+ "epoch": 0.68,
244
+ "grad_norm": 11.456948280334473,
245
+ "learning_rate": 2e-05,
246
+ "loss": 0.7676,
247
+ "step": 68
248
+ },
249
+ {
250
+ "epoch": 0.7,
251
+ "grad_norm": 1.6395078897476196,
252
+ "learning_rate": 2e-05,
253
+ "loss": 0.564,
254
+ "step": 70
255
+ },
256
+ {
257
+ "epoch": 0.72,
258
+ "grad_norm": 2.9087085723876953,
259
+ "learning_rate": 2e-05,
260
+ "loss": 0.5242,
261
+ "step": 72
262
+ },
263
+ {
264
+ "epoch": 0.74,
265
+ "grad_norm": 2.2395269870758057,
266
+ "learning_rate": 2e-05,
267
+ "loss": 0.4575,
268
+ "step": 74
269
+ },
270
+ {
271
+ "epoch": 0.76,
272
+ "grad_norm": 7.380107879638672,
273
+ "learning_rate": 2e-05,
274
+ "loss": 0.5474,
275
+ "step": 76
276
+ },
277
+ {
278
+ "epoch": 0.78,
279
+ "grad_norm": 3.7302591800689697,
280
+ "learning_rate": 2e-05,
281
+ "loss": 0.365,
282
+ "step": 78
283
+ },
284
+ {
285
+ "epoch": 0.8,
286
+ "grad_norm": 7.54459810256958,
287
+ "learning_rate": 2e-05,
288
+ "loss": 0.8384,
289
+ "step": 80
290
+ },
291
+ {
292
+ "epoch": 0.82,
293
+ "grad_norm": 1.8007885217666626,
294
+ "learning_rate": 2e-05,
295
+ "loss": 0.354,
296
+ "step": 82
297
+ },
298
+ {
299
+ "epoch": 0.84,
300
+ "grad_norm": 1.5501184463500977,
301
+ "learning_rate": 2e-05,
302
+ "loss": 0.5158,
303
+ "step": 84
304
+ },
305
+ {
306
+ "epoch": 0.86,
307
+ "grad_norm": 7.850472927093506,
308
+ "learning_rate": 2e-05,
309
+ "loss": 0.5859,
310
+ "step": 86
311
+ },
312
+ {
313
+ "epoch": 0.88,
314
+ "grad_norm": 2.9437716007232666,
315
+ "learning_rate": 2e-05,
316
+ "loss": 0.5186,
317
+ "step": 88
318
+ },
319
+ {
320
+ "epoch": 0.9,
321
+ "grad_norm": 6.130828380584717,
322
+ "learning_rate": 2e-05,
323
+ "loss": 0.5325,
324
+ "step": 90
325
+ },
326
+ {
327
+ "epoch": 0.92,
328
+ "grad_norm": 3.8235013484954834,
329
+ "learning_rate": 2e-05,
330
+ "loss": 0.4294,
331
+ "step": 92
332
+ },
333
+ {
334
+ "epoch": 0.94,
335
+ "grad_norm": 10.231576919555664,
336
+ "learning_rate": 2e-05,
337
+ "loss": 0.5122,
338
+ "step": 94
339
+ },
340
+ {
341
+ "epoch": 0.96,
342
+ "grad_norm": 3.4070258140563965,
343
+ "learning_rate": 2e-05,
344
+ "loss": 0.4346,
345
+ "step": 96
346
+ },
347
+ {
348
+ "epoch": 0.98,
349
+ "grad_norm": 2.635401725769043,
350
+ "learning_rate": 2e-05,
351
+ "loss": 0.4543,
352
+ "step": 98
353
+ },
354
+ {
355
+ "epoch": 1.0,
356
+ "grad_norm": 4.57978630065918,
357
+ "learning_rate": 2e-05,
358
+ "loss": 0.8027,
359
+ "step": 100
360
+ },
361
+ {
362
+ "epoch": 1.0,
363
+ "step": 100,
364
+ "total_flos": 1.6576168127889408e+16,
365
+ "train_loss": 0.5424420166015625,
366
+ "train_runtime": 289.5764,
367
+ "train_samples_per_second": 1.381,
368
+ "train_steps_per_second": 0.345
369
+ }
370
+ ],
371
+ "logging_steps": 2,
372
+ "max_steps": 100,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 1,
375
+ "save_steps": 500,
376
+ "stateful_callbacks": {
377
+ "TrainerControl": {
378
+ "args": {
379
+ "should_epoch_stop": false,
380
+ "should_evaluate": false,
381
+ "should_log": false,
382
+ "should_save": false,
383
+ "should_training_stop": false
384
+ },
385
+ "attributes": {}
386
+ }
387
+ },
388
+ "total_flos": 1.6576168127889408e+16,
389
+ "train_batch_size": 2,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round10.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b16e88793c124816f1f558de280952798c45b5b80104feb95b1ea3c823f4317
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round12.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb99f249d83641dd27f96f403f9a3896cc2c106017f1dac1743ed4462b07ee5
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round15.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1544b776c37e6184161b7dce77cfc588b766260a10846dd63415c96f9147be4
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round17.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34be3c453cd7a2af03cf351083bd8e44c0b7da498f43599dc9da884887ab8ec6
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23b15c520084743ef62975eae5db8d4352e6c8829c36aa7360dc593c57110e2c
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round20.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:adc4c8842a382db73fdc998bf2ec50d04d6f160cab990be0e64eaa17ee6d697d
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c91d31a53873f602e9654cdc5d71ec452143909946552d992ae47d2e138a505
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_client_model_round7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4b6565669e5be448c7713d437c76b999e97c1f563a913e8d7b24fb23f5a026b
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/2_trainer_state.json ADDED
@@ -0,0 +1,392 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 100,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02,
13
+ "grad_norm": 7.872583389282227,
14
+ "learning_rate": 2e-05,
15
+ "loss": 0.4408,
16
+ "step": 2
17
+ },
18
+ {
19
+ "epoch": 0.04,
20
+ "grad_norm": 3.1676270961761475,
21
+ "learning_rate": 2e-05,
22
+ "loss": 0.2183,
23
+ "step": 4
24
+ },
25
+ {
26
+ "epoch": 0.06,
27
+ "grad_norm": 3.10235595703125,
28
+ "learning_rate": 2e-05,
29
+ "loss": 0.3065,
30
+ "step": 6
31
+ },
32
+ {
33
+ "epoch": 0.08,
34
+ "grad_norm": 1.986329436302185,
35
+ "learning_rate": 2e-05,
36
+ "loss": 0.1555,
37
+ "step": 8
38
+ },
39
+ {
40
+ "epoch": 0.1,
41
+ "grad_norm": 6.654890060424805,
42
+ "learning_rate": 2e-05,
43
+ "loss": 0.4976,
44
+ "step": 10
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "grad_norm": 8.408140182495117,
49
+ "learning_rate": 2e-05,
50
+ "loss": 0.6118,
51
+ "step": 12
52
+ },
53
+ {
54
+ "epoch": 0.14,
55
+ "grad_norm": 9.52741813659668,
56
+ "learning_rate": 2e-05,
57
+ "loss": 1.0054,
58
+ "step": 14
59
+ },
60
+ {
61
+ "epoch": 0.16,
62
+ "grad_norm": 3.1671223640441895,
63
+ "learning_rate": 2e-05,
64
+ "loss": 0.3345,
65
+ "step": 16
66
+ },
67
+ {
68
+ "epoch": 0.18,
69
+ "grad_norm": 6.954838275909424,
70
+ "learning_rate": 2e-05,
71
+ "loss": 0.5977,
72
+ "step": 18
73
+ },
74
+ {
75
+ "epoch": 0.2,
76
+ "grad_norm": 3.3944637775421143,
77
+ "learning_rate": 2e-05,
78
+ "loss": 0.3164,
79
+ "step": 20
80
+ },
81
+ {
82
+ "epoch": 0.22,
83
+ "grad_norm": 3.7810606956481934,
84
+ "learning_rate": 2e-05,
85
+ "loss": 0.5396,
86
+ "step": 22
87
+ },
88
+ {
89
+ "epoch": 0.24,
90
+ "grad_norm": 2.8664538860321045,
91
+ "learning_rate": 2e-05,
92
+ "loss": 0.5024,
93
+ "step": 24
94
+ },
95
+ {
96
+ "epoch": 0.26,
97
+ "grad_norm": 5.678687572479248,
98
+ "learning_rate": 2e-05,
99
+ "loss": 0.4766,
100
+ "step": 26
101
+ },
102
+ {
103
+ "epoch": 0.28,
104
+ "grad_norm": 7.057774066925049,
105
+ "learning_rate": 2e-05,
106
+ "loss": 0.4387,
107
+ "step": 28
108
+ },
109
+ {
110
+ "epoch": 0.3,
111
+ "grad_norm": 2.025454044342041,
112
+ "learning_rate": 2e-05,
113
+ "loss": 0.3817,
114
+ "step": 30
115
+ },
116
+ {
117
+ "epoch": 0.32,
118
+ "grad_norm": 2.349515676498413,
119
+ "learning_rate": 2e-05,
120
+ "loss": 0.3013,
121
+ "step": 32
122
+ },
123
+ {
124
+ "epoch": 0.34,
125
+ "grad_norm": 3.751689910888672,
126
+ "learning_rate": 2e-05,
127
+ "loss": 0.3142,
128
+ "step": 34
129
+ },
130
+ {
131
+ "epoch": 0.36,
132
+ "grad_norm": 9.54448127746582,
133
+ "learning_rate": 2e-05,
134
+ "loss": 0.6094,
135
+ "step": 36
136
+ },
137
+ {
138
+ "epoch": 0.38,
139
+ "grad_norm": 2.6551620960235596,
140
+ "learning_rate": 2e-05,
141
+ "loss": 0.3418,
142
+ "step": 38
143
+ },
144
+ {
145
+ "epoch": 0.4,
146
+ "grad_norm": 7.663268089294434,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.3774,
149
+ "step": 40
150
+ },
151
+ {
152
+ "epoch": 0.42,
153
+ "grad_norm": 5.838813781738281,
154
+ "learning_rate": 2e-05,
155
+ "loss": 0.5634,
156
+ "step": 42
157
+ },
158
+ {
159
+ "epoch": 0.44,
160
+ "grad_norm": 3.6920807361602783,
161
+ "learning_rate": 2e-05,
162
+ "loss": 0.1825,
163
+ "step": 44
164
+ },
165
+ {
166
+ "epoch": 0.46,
167
+ "grad_norm": 2.167635202407837,
168
+ "learning_rate": 2e-05,
169
+ "loss": 0.2479,
170
+ "step": 46
171
+ },
172
+ {
173
+ "epoch": 0.48,
174
+ "grad_norm": 1.2298402786254883,
175
+ "learning_rate": 2e-05,
176
+ "loss": 0.274,
177
+ "step": 48
178
+ },
179
+ {
180
+ "epoch": 0.5,
181
+ "grad_norm": 8.415931701660156,
182
+ "learning_rate": 2e-05,
183
+ "loss": 0.6021,
184
+ "step": 50
185
+ },
186
+ {
187
+ "epoch": 0.52,
188
+ "grad_norm": 5.416069984436035,
189
+ "learning_rate": 2e-05,
190
+ "loss": 0.4319,
191
+ "step": 52
192
+ },
193
+ {
194
+ "epoch": 0.54,
195
+ "grad_norm": 2.6783201694488525,
196
+ "learning_rate": 2e-05,
197
+ "loss": 0.4147,
198
+ "step": 54
199
+ },
200
+ {
201
+ "epoch": 0.56,
202
+ "grad_norm": 1.76664137840271,
203
+ "learning_rate": 2e-05,
204
+ "loss": 0.2076,
205
+ "step": 56
206
+ },
207
+ {
208
+ "epoch": 0.58,
209
+ "grad_norm": 3.5323493480682373,
210
+ "learning_rate": 2e-05,
211
+ "loss": 0.3228,
212
+ "step": 58
213
+ },
214
+ {
215
+ "epoch": 0.6,
216
+ "grad_norm": 4.204054832458496,
217
+ "learning_rate": 2e-05,
218
+ "loss": 0.3702,
219
+ "step": 60
220
+ },
221
+ {
222
+ "epoch": 0.62,
223
+ "grad_norm": 16.42522621154785,
224
+ "learning_rate": 2e-05,
225
+ "loss": 0.8672,
226
+ "step": 62
227
+ },
228
+ {
229
+ "epoch": 0.64,
230
+ "grad_norm": 1.9896955490112305,
231
+ "learning_rate": 2e-05,
232
+ "loss": 0.4229,
233
+ "step": 64
234
+ },
235
+ {
236
+ "epoch": 0.66,
237
+ "grad_norm": 6.183014869689941,
238
+ "learning_rate": 2e-05,
239
+ "loss": 0.552,
240
+ "step": 66
241
+ },
242
+ {
243
+ "epoch": 0.68,
244
+ "grad_norm": 5.789742469787598,
245
+ "learning_rate": 2e-05,
246
+ "loss": 0.3384,
247
+ "step": 68
248
+ },
249
+ {
250
+ "epoch": 0.7,
251
+ "grad_norm": 11.22754955291748,
252
+ "learning_rate": 2e-05,
253
+ "loss": 0.3374,
254
+ "step": 70
255
+ },
256
+ {
257
+ "epoch": 0.72,
258
+ "grad_norm": 1.9776259660720825,
259
+ "learning_rate": 2e-05,
260
+ "loss": 0.3358,
261
+ "step": 72
262
+ },
263
+ {
264
+ "epoch": 0.74,
265
+ "grad_norm": 4.03769588470459,
266
+ "learning_rate": 2e-05,
267
+ "loss": 0.4689,
268
+ "step": 74
269
+ },
270
+ {
271
+ "epoch": 0.76,
272
+ "grad_norm": 6.714748382568359,
273
+ "learning_rate": 2e-05,
274
+ "loss": 0.7046,
275
+ "step": 76
276
+ },
277
+ {
278
+ "epoch": 0.78,
279
+ "grad_norm": 1.7425217628479004,
280
+ "learning_rate": 2e-05,
281
+ "loss": 0.0582,
282
+ "step": 78
283
+ },
284
+ {
285
+ "epoch": 0.8,
286
+ "grad_norm": 2.1245808601379395,
287
+ "learning_rate": 2e-05,
288
+ "loss": 0.3812,
289
+ "step": 80
290
+ },
291
+ {
292
+ "epoch": 0.82,
293
+ "grad_norm": 8.465872764587402,
294
+ "learning_rate": 2e-05,
295
+ "loss": 0.4424,
296
+ "step": 82
297
+ },
298
+ {
299
+ "epoch": 0.84,
300
+ "grad_norm": 2.888294219970703,
301
+ "learning_rate": 2e-05,
302
+ "loss": 0.1748,
303
+ "step": 84
304
+ },
305
+ {
306
+ "epoch": 0.86,
307
+ "grad_norm": 3.158863067626953,
308
+ "learning_rate": 2e-05,
309
+ "loss": 0.2654,
310
+ "step": 86
311
+ },
312
+ {
313
+ "epoch": 0.88,
314
+ "grad_norm": 7.765163898468018,
315
+ "learning_rate": 2e-05,
316
+ "loss": 0.448,
317
+ "step": 88
318
+ },
319
+ {
320
+ "epoch": 0.9,
321
+ "grad_norm": 2.6586227416992188,
322
+ "learning_rate": 2e-05,
323
+ "loss": 0.1208,
324
+ "step": 90
325
+ },
326
+ {
327
+ "epoch": 0.92,
328
+ "grad_norm": 4.466168403625488,
329
+ "learning_rate": 2e-05,
330
+ "loss": 0.4543,
331
+ "step": 92
332
+ },
333
+ {
334
+ "epoch": 0.94,
335
+ "grad_norm": 9.527925491333008,
336
+ "learning_rate": 2e-05,
337
+ "loss": 1.0899,
338
+ "step": 94
339
+ },
340
+ {
341
+ "epoch": 0.96,
342
+ "grad_norm": 7.631684303283691,
343
+ "learning_rate": 2e-05,
344
+ "loss": 0.9442,
345
+ "step": 96
346
+ },
347
+ {
348
+ "epoch": 0.98,
349
+ "grad_norm": 9.0326566696167,
350
+ "learning_rate": 2e-05,
351
+ "loss": 0.7216,
352
+ "step": 98
353
+ },
354
+ {
355
+ "epoch": 1.0,
356
+ "grad_norm": 9.811245918273926,
357
+ "learning_rate": 2e-05,
358
+ "loss": 0.5501,
359
+ "step": 100
360
+ },
361
+ {
362
+ "epoch": 1.0,
363
+ "step": 100,
364
+ "total_flos": 1.9115952041885696e+16,
365
+ "train_loss": 0.44124862670898435,
366
+ "train_runtime": 324.4955,
367
+ "train_samples_per_second": 1.233,
368
+ "train_steps_per_second": 0.308
369
+ }
370
+ ],
371
+ "logging_steps": 2,
372
+ "max_steps": 100,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 1,
375
+ "save_steps": 500,
376
+ "stateful_callbacks": {
377
+ "TrainerControl": {
378
+ "args": {
379
+ "should_epoch_stop": false,
380
+ "should_evaluate": false,
381
+ "should_log": false,
382
+ "should_save": false,
383
+ "should_training_stop": false
384
+ },
385
+ "attributes": {}
386
+ }
387
+ },
388
+ "total_flos": 1.9115952041885696e+16,
389
+ "train_batch_size": 2,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round10.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b95720db68933091baf1fdecaa1719d3ae6d74f9d586153dfbdd9a8d412efd5
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round12.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9462aef5d3987e462b4eb54f5767c7a989d765da13e27c013f24ea225359ff85
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round15.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da91a450b0e42f8409cf5f1538d602d5016eb3af4a99a673858d20fac1506bf4
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round17.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2137138ebfad8ba608a66235c099e93ed835a9caccde901a0cce2619288f7d1
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a1463e43df9db510a4857be4caea7caef4417a1d68d2af7784cc2114322d425
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round20.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6986573a8d5a03d8aafbd13449ec4c1b317445289bb1d6fe3f86f5a60b2293a1
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b459889ad8c0dddbdb5b5892a3dc82a8c7aff84f714b2aa33a6d025449e86354
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_client_model_round7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4c6ae5cabef78b40781076f85a494852e0479f8117956282442c035a9696f6c
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/3_trainer_state.json ADDED
@@ -0,0 +1,392 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 100,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02,
13
+ "grad_norm": 4.6497626304626465,
14
+ "learning_rate": 2e-05,
15
+ "loss": 0.5825,
16
+ "step": 2
17
+ },
18
+ {
19
+ "epoch": 0.04,
20
+ "grad_norm": 3.6290929317474365,
21
+ "learning_rate": 2e-05,
22
+ "loss": 0.915,
23
+ "step": 4
24
+ },
25
+ {
26
+ "epoch": 0.06,
27
+ "grad_norm": 2.3345847129821777,
28
+ "learning_rate": 2e-05,
29
+ "loss": 0.4902,
30
+ "step": 6
31
+ },
32
+ {
33
+ "epoch": 0.08,
34
+ "grad_norm": 5.3167009353637695,
35
+ "learning_rate": 2e-05,
36
+ "loss": 0.4688,
37
+ "step": 8
38
+ },
39
+ {
40
+ "epoch": 0.1,
41
+ "grad_norm": 3.8224802017211914,
42
+ "learning_rate": 2e-05,
43
+ "loss": 0.5308,
44
+ "step": 10
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "grad_norm": 2.823826551437378,
49
+ "learning_rate": 2e-05,
50
+ "loss": 0.2563,
51
+ "step": 12
52
+ },
53
+ {
54
+ "epoch": 0.14,
55
+ "grad_norm": 2.9755983352661133,
56
+ "learning_rate": 2e-05,
57
+ "loss": 0.3347,
58
+ "step": 14
59
+ },
60
+ {
61
+ "epoch": 0.16,
62
+ "grad_norm": 2.022836923599243,
63
+ "learning_rate": 2e-05,
64
+ "loss": 0.426,
65
+ "step": 16
66
+ },
67
+ {
68
+ "epoch": 0.18,
69
+ "grad_norm": 1.7052284479141235,
70
+ "learning_rate": 2e-05,
71
+ "loss": 0.5396,
72
+ "step": 18
73
+ },
74
+ {
75
+ "epoch": 0.2,
76
+ "grad_norm": 1.3424859046936035,
77
+ "learning_rate": 2e-05,
78
+ "loss": 0.4526,
79
+ "step": 20
80
+ },
81
+ {
82
+ "epoch": 0.22,
83
+ "grad_norm": 5.240848541259766,
84
+ "learning_rate": 2e-05,
85
+ "loss": 0.4441,
86
+ "step": 22
87
+ },
88
+ {
89
+ "epoch": 0.24,
90
+ "grad_norm": 4.661110877990723,
91
+ "learning_rate": 2e-05,
92
+ "loss": 0.6226,
93
+ "step": 24
94
+ },
95
+ {
96
+ "epoch": 0.26,
97
+ "grad_norm": 1.6846646070480347,
98
+ "learning_rate": 2e-05,
99
+ "loss": 0.3237,
100
+ "step": 26
101
+ },
102
+ {
103
+ "epoch": 0.28,
104
+ "grad_norm": 5.903859615325928,
105
+ "learning_rate": 2e-05,
106
+ "loss": 0.7686,
107
+ "step": 28
108
+ },
109
+ {
110
+ "epoch": 0.3,
111
+ "grad_norm": 2.1405582427978516,
112
+ "learning_rate": 2e-05,
113
+ "loss": 0.6973,
114
+ "step": 30
115
+ },
116
+ {
117
+ "epoch": 0.32,
118
+ "grad_norm": 3.7082715034484863,
119
+ "learning_rate": 2e-05,
120
+ "loss": 0.5305,
121
+ "step": 32
122
+ },
123
+ {
124
+ "epoch": 0.34,
125
+ "grad_norm": 4.9589338302612305,
126
+ "learning_rate": 2e-05,
127
+ "loss": 0.4478,
128
+ "step": 34
129
+ },
130
+ {
131
+ "epoch": 0.36,
132
+ "grad_norm": 7.48822546005249,
133
+ "learning_rate": 2e-05,
134
+ "loss": 0.4746,
135
+ "step": 36
136
+ },
137
+ {
138
+ "epoch": 0.38,
139
+ "grad_norm": 8.993917465209961,
140
+ "learning_rate": 2e-05,
141
+ "loss": 0.501,
142
+ "step": 38
143
+ },
144
+ {
145
+ "epoch": 0.4,
146
+ "grad_norm": 2.639606237411499,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.3493,
149
+ "step": 40
150
+ },
151
+ {
152
+ "epoch": 0.42,
153
+ "grad_norm": 7.567841529846191,
154
+ "learning_rate": 2e-05,
155
+ "loss": 0.5112,
156
+ "step": 42
157
+ },
158
+ {
159
+ "epoch": 0.44,
160
+ "grad_norm": 3.463698625564575,
161
+ "learning_rate": 2e-05,
162
+ "loss": 0.3148,
163
+ "step": 44
164
+ },
165
+ {
166
+ "epoch": 0.46,
167
+ "grad_norm": 3.490788698196411,
168
+ "learning_rate": 2e-05,
169
+ "loss": 1.1011,
170
+ "step": 46
171
+ },
172
+ {
173
+ "epoch": 0.48,
174
+ "grad_norm": 3.9706830978393555,
175
+ "learning_rate": 2e-05,
176
+ "loss": 0.303,
177
+ "step": 48
178
+ },
179
+ {
180
+ "epoch": 0.5,
181
+ "grad_norm": 4.992152214050293,
182
+ "learning_rate": 2e-05,
183
+ "loss": 0.6089,
184
+ "step": 50
185
+ },
186
+ {
187
+ "epoch": 0.52,
188
+ "grad_norm": 1.5260461568832397,
189
+ "learning_rate": 2e-05,
190
+ "loss": 0.1576,
191
+ "step": 52
192
+ },
193
+ {
194
+ "epoch": 0.54,
195
+ "grad_norm": 2.76921010017395,
196
+ "learning_rate": 2e-05,
197
+ "loss": 0.4696,
198
+ "step": 54
199
+ },
200
+ {
201
+ "epoch": 0.56,
202
+ "grad_norm": 3.4018092155456543,
203
+ "learning_rate": 2e-05,
204
+ "loss": 0.5396,
205
+ "step": 56
206
+ },
207
+ {
208
+ "epoch": 0.58,
209
+ "grad_norm": 3.256915330886841,
210
+ "learning_rate": 2e-05,
211
+ "loss": 0.3384,
212
+ "step": 58
213
+ },
214
+ {
215
+ "epoch": 0.6,
216
+ "grad_norm": 2.284102439880371,
217
+ "learning_rate": 2e-05,
218
+ "loss": 0.2712,
219
+ "step": 60
220
+ },
221
+ {
222
+ "epoch": 0.62,
223
+ "grad_norm": 5.065925121307373,
224
+ "learning_rate": 2e-05,
225
+ "loss": 0.3423,
226
+ "step": 62
227
+ },
228
+ {
229
+ "epoch": 0.64,
230
+ "grad_norm": 2.2383081912994385,
231
+ "learning_rate": 2e-05,
232
+ "loss": 0.5986,
233
+ "step": 64
234
+ },
235
+ {
236
+ "epoch": 0.66,
237
+ "grad_norm": 3.604736566543579,
238
+ "learning_rate": 2e-05,
239
+ "loss": 0.594,
240
+ "step": 66
241
+ },
242
+ {
243
+ "epoch": 0.68,
244
+ "grad_norm": 2.073974609375,
245
+ "learning_rate": 2e-05,
246
+ "loss": 0.1852,
247
+ "step": 68
248
+ },
249
+ {
250
+ "epoch": 0.7,
251
+ "grad_norm": 4.072497367858887,
252
+ "learning_rate": 2e-05,
253
+ "loss": 0.3573,
254
+ "step": 70
255
+ },
256
+ {
257
+ "epoch": 0.72,
258
+ "grad_norm": 2.104954719543457,
259
+ "learning_rate": 2e-05,
260
+ "loss": 0.4517,
261
+ "step": 72
262
+ },
263
+ {
264
+ "epoch": 0.74,
265
+ "grad_norm": 20.248361587524414,
266
+ "learning_rate": 2e-05,
267
+ "loss": 0.5629,
268
+ "step": 74
269
+ },
270
+ {
271
+ "epoch": 0.76,
272
+ "grad_norm": 5.614309787750244,
273
+ "learning_rate": 2e-05,
274
+ "loss": 0.5081,
275
+ "step": 76
276
+ },
277
+ {
278
+ "epoch": 0.78,
279
+ "grad_norm": 3.6506497859954834,
280
+ "learning_rate": 2e-05,
281
+ "loss": 0.7004,
282
+ "step": 78
283
+ },
284
+ {
285
+ "epoch": 0.8,
286
+ "grad_norm": 0.7987788319587708,
287
+ "learning_rate": 2e-05,
288
+ "loss": 0.2193,
289
+ "step": 80
290
+ },
291
+ {
292
+ "epoch": 0.82,
293
+ "grad_norm": 2.1284451484680176,
294
+ "learning_rate": 2e-05,
295
+ "loss": 0.3389,
296
+ "step": 82
297
+ },
298
+ {
299
+ "epoch": 0.84,
300
+ "grad_norm": 7.366853713989258,
301
+ "learning_rate": 2e-05,
302
+ "loss": 0.748,
303
+ "step": 84
304
+ },
305
+ {
306
+ "epoch": 0.86,
307
+ "grad_norm": 4.810544490814209,
308
+ "learning_rate": 2e-05,
309
+ "loss": 0.7964,
310
+ "step": 86
311
+ },
312
+ {
313
+ "epoch": 0.88,
314
+ "grad_norm": 6.225423336029053,
315
+ "learning_rate": 2e-05,
316
+ "loss": 0.5322,
317
+ "step": 88
318
+ },
319
+ {
320
+ "epoch": 0.9,
321
+ "grad_norm": 4.1773576736450195,
322
+ "learning_rate": 2e-05,
323
+ "loss": 0.5176,
324
+ "step": 90
325
+ },
326
+ {
327
+ "epoch": 0.92,
328
+ "grad_norm": 6.231110572814941,
329
+ "learning_rate": 2e-05,
330
+ "loss": 0.4377,
331
+ "step": 92
332
+ },
333
+ {
334
+ "epoch": 0.94,
335
+ "grad_norm": 2.0640790462493896,
336
+ "learning_rate": 2e-05,
337
+ "loss": 0.3445,
338
+ "step": 94
339
+ },
340
+ {
341
+ "epoch": 0.96,
342
+ "grad_norm": 3.780561685562134,
343
+ "learning_rate": 2e-05,
344
+ "loss": 0.5642,
345
+ "step": 96
346
+ },
347
+ {
348
+ "epoch": 0.98,
349
+ "grad_norm": 4.6030449867248535,
350
+ "learning_rate": 2e-05,
351
+ "loss": 0.363,
352
+ "step": 98
353
+ },
354
+ {
355
+ "epoch": 1.0,
356
+ "grad_norm": 1.9157155752182007,
357
+ "learning_rate": 2e-05,
358
+ "loss": 0.2215,
359
+ "step": 100
360
+ },
361
+ {
362
+ "epoch": 1.0,
363
+ "step": 100,
364
+ "total_flos": 2.386061380550656e+16,
365
+ "train_loss": 0.4830999755859375,
366
+ "train_runtime": 409.3895,
367
+ "train_samples_per_second": 0.977,
368
+ "train_steps_per_second": 0.244
369
+ }
370
+ ],
371
+ "logging_steps": 2,
372
+ "max_steps": 100,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 1,
375
+ "save_steps": 500,
376
+ "stateful_callbacks": {
377
+ "TrainerControl": {
378
+ "args": {
379
+ "should_epoch_stop": false,
380
+ "should_evaluate": false,
381
+ "should_log": false,
382
+ "should_save": false,
383
+ "should_training_stop": false
384
+ },
385
+ "attributes": {}
386
+ }
387
+ },
388
+ "total_flos": 2.386061380550656e+16,
389
+ "train_batch_size": 2,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round10.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:893681346ba343e9d0fa2f36e51accfc6d218f0664ecca940e57f3d2dc374ad9
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round12.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b46411ee8d7334f558926aaf247e2004138dcd997e035c87686449bcfc1cc469
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round15.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bdb196c7e4df84cc6052571a2d10021bc4e6208c3ed2b79af86450bb25dee17c
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round17.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b45484c94bb3ab612febd88336db3b78e201867c1e805bbd435f37ef49104931
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:118254270d48a1bd72d800e0b757f7eee2d6879af1e323d473ca3b25a5b156fe
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round20.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:479108ba00e734b040fe1210bcbd5d58bcbf9b9d213843eada9ace7329f09763
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5cda0740f3777ae9d6bc03793b0f525db6fc40288d9de12a48092382737c4350
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_client_model_round7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3def9c46010f15f880769a9efe2296b8114ce3d0a5ad9754de24319c2fe578c
3
+ size 395786922
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/4_trainer_state.json ADDED
@@ -0,0 +1,392 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 100,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02,
13
+ "grad_norm": 3.530467987060547,
14
+ "learning_rate": 2e-05,
15
+ "loss": 0.2157,
16
+ "step": 2
17
+ },
18
+ {
19
+ "epoch": 0.04,
20
+ "grad_norm": 1.8090039491653442,
21
+ "learning_rate": 2e-05,
22
+ "loss": 0.1584,
23
+ "step": 4
24
+ },
25
+ {
26
+ "epoch": 0.06,
27
+ "grad_norm": 2.7021498680114746,
28
+ "learning_rate": 2e-05,
29
+ "loss": 0.2725,
30
+ "step": 6
31
+ },
32
+ {
33
+ "epoch": 0.08,
34
+ "grad_norm": 1.0994998216629028,
35
+ "learning_rate": 2e-05,
36
+ "loss": 0.1925,
37
+ "step": 8
38
+ },
39
+ {
40
+ "epoch": 0.1,
41
+ "grad_norm": 4.899983882904053,
42
+ "learning_rate": 2e-05,
43
+ "loss": 0.4641,
44
+ "step": 10
45
+ },
46
+ {
47
+ "epoch": 0.12,
48
+ "grad_norm": 1.8486827611923218,
49
+ "learning_rate": 2e-05,
50
+ "loss": 0.6014,
51
+ "step": 12
52
+ },
53
+ {
54
+ "epoch": 0.14,
55
+ "grad_norm": 0.669780433177948,
56
+ "learning_rate": 2e-05,
57
+ "loss": 0.1162,
58
+ "step": 14
59
+ },
60
+ {
61
+ "epoch": 0.16,
62
+ "grad_norm": 4.612163066864014,
63
+ "learning_rate": 2e-05,
64
+ "loss": 0.5449,
65
+ "step": 16
66
+ },
67
+ {
68
+ "epoch": 0.18,
69
+ "grad_norm": 1.251813292503357,
70
+ "learning_rate": 2e-05,
71
+ "loss": 0.3157,
72
+ "step": 18
73
+ },
74
+ {
75
+ "epoch": 0.2,
76
+ "grad_norm": 4.452538013458252,
77
+ "learning_rate": 2e-05,
78
+ "loss": 0.1715,
79
+ "step": 20
80
+ },
81
+ {
82
+ "epoch": 0.22,
83
+ "grad_norm": 4.419332981109619,
84
+ "learning_rate": 2e-05,
85
+ "loss": 0.2881,
86
+ "step": 22
87
+ },
88
+ {
89
+ "epoch": 0.24,
90
+ "grad_norm": 2.8015778064727783,
91
+ "learning_rate": 2e-05,
92
+ "loss": 0.8685,
93
+ "step": 24
94
+ },
95
+ {
96
+ "epoch": 0.26,
97
+ "grad_norm": 5.626010894775391,
98
+ "learning_rate": 2e-05,
99
+ "loss": 1.0619,
100
+ "step": 26
101
+ },
102
+ {
103
+ "epoch": 0.28,
104
+ "grad_norm": 7.333406925201416,
105
+ "learning_rate": 2e-05,
106
+ "loss": 1.0535,
107
+ "step": 28
108
+ },
109
+ {
110
+ "epoch": 0.3,
111
+ "grad_norm": 6.450397968292236,
112
+ "learning_rate": 2e-05,
113
+ "loss": 0.9023,
114
+ "step": 30
115
+ },
116
+ {
117
+ "epoch": 0.32,
118
+ "grad_norm": 3.454080820083618,
119
+ "learning_rate": 2e-05,
120
+ "loss": 0.4188,
121
+ "step": 32
122
+ },
123
+ {
124
+ "epoch": 0.34,
125
+ "grad_norm": 2.1945369243621826,
126
+ "learning_rate": 2e-05,
127
+ "loss": 0.4095,
128
+ "step": 34
129
+ },
130
+ {
131
+ "epoch": 0.36,
132
+ "grad_norm": 2.2680203914642334,
133
+ "learning_rate": 2e-05,
134
+ "loss": 0.9443,
135
+ "step": 36
136
+ },
137
+ {
138
+ "epoch": 0.38,
139
+ "grad_norm": 5.364918231964111,
140
+ "learning_rate": 2e-05,
141
+ "loss": 0.2831,
142
+ "step": 38
143
+ },
144
+ {
145
+ "epoch": 0.4,
146
+ "grad_norm": 5.871020793914795,
147
+ "learning_rate": 2e-05,
148
+ "loss": 0.4717,
149
+ "step": 40
150
+ },
151
+ {
152
+ "epoch": 0.42,
153
+ "grad_norm": 0.05804154649376869,
154
+ "learning_rate": 2e-05,
155
+ "loss": 0.2073,
156
+ "step": 42
157
+ },
158
+ {
159
+ "epoch": 0.44,
160
+ "grad_norm": 3.274716854095459,
161
+ "learning_rate": 2e-05,
162
+ "loss": 0.595,
163
+ "step": 44
164
+ },
165
+ {
166
+ "epoch": 0.46,
167
+ "grad_norm": 2.444748878479004,
168
+ "learning_rate": 2e-05,
169
+ "loss": 0.395,
170
+ "step": 46
171
+ },
172
+ {
173
+ "epoch": 0.48,
174
+ "grad_norm": 2.9482669830322266,
175
+ "learning_rate": 2e-05,
176
+ "loss": 0.3739,
177
+ "step": 48
178
+ },
179
+ {
180
+ "epoch": 0.5,
181
+ "grad_norm": 4.834102630615234,
182
+ "learning_rate": 2e-05,
183
+ "loss": 0.464,
184
+ "step": 50
185
+ },
186
+ {
187
+ "epoch": 0.52,
188
+ "grad_norm": 4.0864577293396,
189
+ "learning_rate": 2e-05,
190
+ "loss": 1.0002,
191
+ "step": 52
192
+ },
193
+ {
194
+ "epoch": 0.54,
195
+ "grad_norm": 0.2608688473701477,
196
+ "learning_rate": 2e-05,
197
+ "loss": 0.3152,
198
+ "step": 54
199
+ },
200
+ {
201
+ "epoch": 0.56,
202
+ "grad_norm": 1.6589609384536743,
203
+ "learning_rate": 2e-05,
204
+ "loss": 0.1979,
205
+ "step": 56
206
+ },
207
+ {
208
+ "epoch": 0.58,
209
+ "grad_norm": 2.3401970863342285,
210
+ "learning_rate": 2e-05,
211
+ "loss": 0.3342,
212
+ "step": 58
213
+ },
214
+ {
215
+ "epoch": 0.6,
216
+ "grad_norm": 2.3094046115875244,
217
+ "learning_rate": 2e-05,
218
+ "loss": 0.2121,
219
+ "step": 60
220
+ },
221
+ {
222
+ "epoch": 0.62,
223
+ "grad_norm": 2.6082193851470947,
224
+ "learning_rate": 2e-05,
225
+ "loss": 0.345,
226
+ "step": 62
227
+ },
228
+ {
229
+ "epoch": 0.64,
230
+ "grad_norm": 7.014457702636719,
231
+ "learning_rate": 2e-05,
232
+ "loss": 0.8164,
233
+ "step": 64
234
+ },
235
+ {
236
+ "epoch": 0.66,
237
+ "grad_norm": 4.024754047393799,
238
+ "learning_rate": 2e-05,
239
+ "loss": 0.7554,
240
+ "step": 66
241
+ },
242
+ {
243
+ "epoch": 0.68,
244
+ "grad_norm": 0.4044950008392334,
245
+ "learning_rate": 2e-05,
246
+ "loss": 0.0986,
247
+ "step": 68
248
+ },
249
+ {
250
+ "epoch": 0.7,
251
+ "grad_norm": 4.961355209350586,
252
+ "learning_rate": 2e-05,
253
+ "loss": 0.4958,
254
+ "step": 70
255
+ },
256
+ {
257
+ "epoch": 0.72,
258
+ "grad_norm": 7.53846549987793,
259
+ "learning_rate": 2e-05,
260
+ "loss": 0.5591,
261
+ "step": 72
262
+ },
263
+ {
264
+ "epoch": 0.74,
265
+ "grad_norm": 0.829863429069519,
266
+ "learning_rate": 2e-05,
267
+ "loss": 0.4915,
268
+ "step": 74
269
+ },
270
+ {
271
+ "epoch": 0.76,
272
+ "grad_norm": 3.274028778076172,
273
+ "learning_rate": 2e-05,
274
+ "loss": 0.4124,
275
+ "step": 76
276
+ },
277
+ {
278
+ "epoch": 0.78,
279
+ "grad_norm": 5.568228721618652,
280
+ "learning_rate": 2e-05,
281
+ "loss": 0.3262,
282
+ "step": 78
283
+ },
284
+ {
285
+ "epoch": 0.8,
286
+ "grad_norm": 6.694833755493164,
287
+ "learning_rate": 2e-05,
288
+ "loss": 0.6133,
289
+ "step": 80
290
+ },
291
+ {
292
+ "epoch": 0.82,
293
+ "grad_norm": 5.746543884277344,
294
+ "learning_rate": 2e-05,
295
+ "loss": 0.8642,
296
+ "step": 82
297
+ },
298
+ {
299
+ "epoch": 0.84,
300
+ "grad_norm": 11.073870658874512,
301
+ "learning_rate": 2e-05,
302
+ "loss": 0.833,
303
+ "step": 84
304
+ },
305
+ {
306
+ "epoch": 0.86,
307
+ "grad_norm": 4.428623676300049,
308
+ "learning_rate": 2e-05,
309
+ "loss": 0.333,
310
+ "step": 86
311
+ },
312
+ {
313
+ "epoch": 0.88,
314
+ "grad_norm": 7.5453033447265625,
315
+ "learning_rate": 2e-05,
316
+ "loss": 0.7605,
317
+ "step": 88
318
+ },
319
+ {
320
+ "epoch": 0.9,
321
+ "grad_norm": 7.614424228668213,
322
+ "learning_rate": 2e-05,
323
+ "loss": 0.2277,
324
+ "step": 90
325
+ },
326
+ {
327
+ "epoch": 0.92,
328
+ "grad_norm": 8.295862197875977,
329
+ "learning_rate": 2e-05,
330
+ "loss": 0.9114,
331
+ "step": 92
332
+ },
333
+ {
334
+ "epoch": 0.94,
335
+ "grad_norm": 2.497485876083374,
336
+ "learning_rate": 2e-05,
337
+ "loss": 0.2709,
338
+ "step": 94
339
+ },
340
+ {
341
+ "epoch": 0.96,
342
+ "grad_norm": 1.357002854347229,
343
+ "learning_rate": 2e-05,
344
+ "loss": 0.6364,
345
+ "step": 96
346
+ },
347
+ {
348
+ "epoch": 0.98,
349
+ "grad_norm": 6.519164562225342,
350
+ "learning_rate": 2e-05,
351
+ "loss": 0.3441,
352
+ "step": 98
353
+ },
354
+ {
355
+ "epoch": 1.0,
356
+ "grad_norm": 2.9073774814605713,
357
+ "learning_rate": 2e-05,
358
+ "loss": 0.4589,
359
+ "step": 100
360
+ },
361
+ {
362
+ "epoch": 1.0,
363
+ "step": 100,
364
+ "total_flos": 2.410703500397773e+16,
365
+ "train_loss": 0.48806270599365237,
366
+ "train_runtime": 603.2643,
367
+ "train_samples_per_second": 0.663,
368
+ "train_steps_per_second": 0.166
369
+ }
370
+ ],
371
+ "logging_steps": 2,
372
+ "max_steps": 100,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 1,
375
+ "save_steps": 500,
376
+ "stateful_callbacks": {
377
+ "TrainerControl": {
378
+ "args": {
379
+ "should_epoch_stop": false,
380
+ "should_evaluate": false,
381
+ "should_log": false,
382
+ "should_save": false,
383
+ "should_training_stop": false
384
+ },
385
+ "attributes": {}
386
+ }
387
+ },
388
+ "total_flos": 2.410703500397773e+16,
389
+ "train_batch_size": 2,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round10.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e98397f8ee641afb7450a1f3f2b7b46bc3b40c3af558ed2250763f6daf8926a1
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round12.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1182ee1ea28f8787bca578e777f053b2ee0a10bd3430a546fb68a7cdcf699d7
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round15.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63ba0332da3f84fc4e4f6fd1d7367267ccf7e78e25877de4de3f58a1c5482730
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round17.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:423c7b5ce9ea433b1ebd72c1011c1ce8ca96c7bc1ca28b81212c078e95829b52
3
+ size 395787774
client_states_fedMultipqfullfreeze_homoAgg_pca_bs4_saveoptim_lr2e-5_5e-5_sc132_4tasks_5rounds_fixitr100_T0125_decay099/5_client_model_round2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bde583558a09eb45715c7e129609121fa1addebd7bba6f8de5847ea26e0ecadd
3
+ size 395786922