dayiethan commited on
Commit
535a908
·
verified ·
1 Parent(s): 7d990a7

Add verified ALTER simulation artifacts (through file 69)

Browse files
Files changed (29) hide show
  1. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_m_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_m_stats.pkl +3 -0
  2. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_m_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_m_step400000.pt +3 -0
  3. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_s_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_s_stats.pkl +3 -0
  4. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_s_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_s_step300000.pt +3 -0
  5. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_xs_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_xs_stats.pkl +3 -0
  6. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_xs_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_xs_step100000.pt +3 -0
  7. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260912_coord_wd1e4_control/twoarm_forwardonly_capacity_coord_m_5permode_steps900000_seed0/twoarm_forwardonly_capacity_coord_m_stats.pkl +3 -0
  8. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260912_coord_wd1e4_control/twoarm_forwardonly_capacity_coord_m_5permode_steps900000_seed0/twoarm_forwardonly_capacity_coord_m_step200000.pt +3 -0
  9. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260912_coord_wd1e4_control/twoarm_forwardonly_capacity_coord_s_5permode_steps900000_seed0/twoarm_forwardonly_capacity_coord_s_stats.pkl +3 -0
  10. data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260912_coord_wd1e4_control/twoarm_forwardonly_capacity_coord_s_5permode_steps900000_seed0/twoarm_forwardonly_capacity_coord_s_step100000.pt +3 -0
  11. data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_10permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step400000.pt +3 -0
  12. data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_10permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step400000_stats.pkl +3 -0
  13. data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_15permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step500000.pt +3 -0
  14. data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_15permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step500000_stats.pkl +3 -0
  15. data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_5permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step200000.pt +3 -0
  16. data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_5permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step200000_stats.pkl +3 -0
  17. data_primary/results/placewipe_current_forwardonly_frozen_base_native_source200_seed1000_1049_rf15/evaluation.json +44 -0
  18. data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-m-fs/full_horizon1800/step400000_20260911/fromscratch-shoulderonly_placewipe_resnet18_rf15/eval_20260911_173643.json +0 -0
  19. data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-m-fs/native_source_retention200/step400000_20260911/evaluation.json +44 -0
  20. data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-s-fs/full_horizon1800/step300000_20260911/fromscratch-shoulderonly_placewipe_resnet18_rf15/eval_20260911_173645.json +0 -0
  21. data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-s-fs/native_source_retention200/step300000_20260911/evaluation.json +44 -0
  22. data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-xs-fs/earlystop_full_validation_selected_horizon1800/step100000_20260911/fromscratch-shoulderonly_placewipe_resnet18_rf15/eval_20260911_173644.json +0 -0
  23. data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-xs-fs/native_source_retention200/step100000_20260911/evaluation.json +44 -0
  24. data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-m-coord/provisional_full_20260913/step200000/native_source_retention200/evaluation.json +44 -0
  25. data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-m-coord/provisional_full_20260913/step200000/twoarm_full200/coord-shoulderonly_placewipe_resnet18_rf15/eval_20260913_173448.json +0 -0
  26. data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-s-coord/provisional_full_20260913/step100000/native_source_retention200/evaluation.json +44 -0
  27. data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-s-coord/provisional_full_20260913/step100000/twoarm_full200/coord-shoulderonly_placewipe_resnet18_rf15/eval_20260913_173446.json +0 -0
  28. data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-xs-coord/provisional_full_20260913/step100000/native_source_retention200/evaluation.json +44 -0
  29. data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-xs-coord/provisional_full_20260913/step100000/twoarm_full200/coord-shoulderonly_placewipe_resnet18_rf15/eval_20260913_173445.json +0 -0
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_m_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_m_stats.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:449f5004b16cb97c88465537ba50f45dac5a70d44532a86b876129bd9a2e8daf
3
+ size 137631
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_m_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_m_step400000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf375c1795cbd29c176a7a2b19d197f90a22b0f6ae96c27ade226489dcd34f53
3
+ size 166674154
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_s_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_s_stats.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1f4f26823460236c9dad25127024b2b498180a1dd0eb85aa1d1476cb2d3771a
3
+ size 137631
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_s_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_s_step300000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ec18da2b4c69322e782708c5a03cbe2d18d6aff7e67432fd05b0a64dcb27488
3
+ size 152770282
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_xs_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_xs_stats.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:523ba77733b6e590d28f2dbc9438711326beab1c564993fa0cc034af23e23975
3
+ size 137640
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260909_attempt2/twoarm_forwardonly_capacity_fs_xs_5permode_steps900000_seed0/twoarm_forwardonly_capacity_fs_xs_step100000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db1dc195cc7eeeafe8024a63d80cddc514e25dd4ebfc59956ccf119e44c360e0
3
+ size 143206178
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260912_coord_wd1e4_control/twoarm_forwardonly_capacity_coord_m_5permode_steps900000_seed0/twoarm_forwardonly_capacity_coord_m_stats.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa7131a7473160cbed352c5682f5ecdf6a167b23453feddf3b29fd92506fca53
3
+ size 32989
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260912_coord_wd1e4_control/twoarm_forwardonly_capacity_coord_m_5permode_steps900000_seed0/twoarm_forwardonly_capacity_coord_m_step200000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3db570858d70bb17cdd5268eea5a3ac26ea6bf5217ee0811bbd75b2b64ba3a61
3
+ size 27085452
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260912_coord_wd1e4_control/twoarm_forwardonly_capacity_coord_s_5permode_steps900000_seed0/twoarm_forwardonly_capacity_coord_s_stats.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a83715569cd208e33f1a74b67a3cbb170a3cdf2686209d985ae4acd1cc092b0
3
+ size 32989
data_primary/checkpoints/arm/twoarm_forwardonly_capacity_ablation_20260912_coord_wd1e4_control/twoarm_forwardonly_capacity_coord_s_5permode_steps900000_seed0/twoarm_forwardonly_capacity_coord_s_step100000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e6b329855a612842b76851b01f41b8af7789668d0ac417f8e49f272a29e3bbe
3
+ size 13178508
data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_10permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step400000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:625a0fafdbc134a3d426007c4c3cb55d4ab8e0857155959be620fbe976f82e87
3
+ size 139722874
data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_10permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step400000_stats.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fc39976025da9f4c40fec6290c409a7b5a2d0c03b9f8f02b6ab95e373e94bb7
3
+ size 52136
data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_15permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step500000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b2f473b340669c9d32b9ded2c93c4d21762062e18cdbc635d9cabee7f60ced1
3
+ size 139722874
data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_15permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step500000_stats.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e709bde44ca4e28fd275dd57f82f59d8495fb6459b55fe773f45a68763912965
3
+ size 74920
data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_5permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step200000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23c2d9a88135191d659121473a458af455615cfab1aeca1331266ae9ef8ee729
3
+ size 139722874
data_primary/checkpoints/pure_full_policy_twoarm/twoarm_placewipe_pure_finetune_5permode_steps900000_seed0/twoarm_placewipe_pure_finetune_step200000_stats.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24e52e7005be8184701fc4710472488a6fb2c14426cc18c1d04434baa22aa652
3
+ size 29354
data_primary/results/placewipe_current_forwardonly_frozen_base_native_source200_seed1000_1049_rf15/evaluation.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "complete": true,
3
+ "contract_path": "artifact://data_primary/results/placewipe_current_forwardonly_frozen_base_native_source200_seed1000_1049_rf15/evaluation_contract.json",
4
+ "duplicate_rollouts": 0,
5
+ "expected_rollouts": 200,
6
+ "ledger_path": "artifact://data_primary/results/placewipe_current_forwardonly_frozen_base_native_source200_seed1000_1049_rf15/episode_ledger.json",
7
+ "method": "base",
8
+ "missing_rollouts": 0,
9
+ "observed_rollouts": 200,
10
+ "panel_type": "source_retention",
11
+ "protocol": "twoarm_native_source200",
12
+ "rollouts_path": "artifact://data_primary/results/placewipe_current_forwardonly_frozen_base_native_source200_seed1000_1049_rf15/rollouts.jsonl",
13
+ "schema": "sealed_evaluation_result.v1",
14
+ "summary": {
15
+ "by_task_mode": {
16
+ "place_return:mode0": {
17
+ "success": 50,
18
+ "success_rate": 1.0,
19
+ "total": 50
20
+ },
21
+ "place_return:mode1": {
22
+ "success": 50,
23
+ "success_rate": 1.0,
24
+ "total": 50
25
+ },
26
+ "wipe:mode0": {
27
+ "success": 45,
28
+ "success_rate": 0.9,
29
+ "total": 50
30
+ },
31
+ "wipe:mode1": {
32
+ "success": 49,
33
+ "success_rate": 0.98,
34
+ "total": 50
35
+ }
36
+ },
37
+ "overall": {
38
+ "success": 194,
39
+ "success_rate": 0.97,
40
+ "total": 200
41
+ }
42
+ },
43
+ "task_family": "two_arm_placewipe"
44
+ }
data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-m-fs/full_horizon1800/step400000_20260911/fromscratch-shoulderonly_placewipe_resnet18_rf15/eval_20260911_173643.json ADDED
The diff for this file is too large to render. See raw diff
 
data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-m-fs/native_source_retention200/step400000_20260911/evaluation.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "complete": true,
3
+ "contract_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-m-fs/native_source_retention200/step400000_20260911/evaluation_contract.json",
4
+ "duplicate_rollouts": 0,
5
+ "expected_rollouts": 200,
6
+ "ledger_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-m-fs/native_source_retention200/step400000_20260911/episode_ledger.json",
7
+ "method": "fs",
8
+ "missing_rollouts": 0,
9
+ "observed_rollouts": 200,
10
+ "panel_type": "source_retention",
11
+ "protocol": "twoarm_native_source200",
12
+ "rollouts_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-m-fs/native_source_retention200/step400000_20260911/rollouts.jsonl",
13
+ "schema": "sealed_evaluation_result.v1",
14
+ "summary": {
15
+ "by_task_mode": {
16
+ "place_return:mode0": {
17
+ "success": 40,
18
+ "success_rate": 0.8,
19
+ "total": 50
20
+ },
21
+ "place_return:mode1": {
22
+ "success": 30,
23
+ "success_rate": 0.6,
24
+ "total": 50
25
+ },
26
+ "wipe:mode0": {
27
+ "success": 3,
28
+ "success_rate": 0.06,
29
+ "total": 50
30
+ },
31
+ "wipe:mode1": {
32
+ "success": 6,
33
+ "success_rate": 0.12,
34
+ "total": 50
35
+ }
36
+ },
37
+ "overall": {
38
+ "success": 79,
39
+ "success_rate": 0.395,
40
+ "total": 200
41
+ }
42
+ },
43
+ "task_family": "two_arm_placewipe"
44
+ }
data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-s-fs/full_horizon1800/step300000_20260911/fromscratch-shoulderonly_placewipe_resnet18_rf15/eval_20260911_173645.json ADDED
The diff for this file is too large to render. See raw diff
 
data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-s-fs/native_source_retention200/step300000_20260911/evaluation.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "complete": true,
3
+ "contract_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-s-fs/native_source_retention200/step300000_20260911/evaluation_contract.json",
4
+ "duplicate_rollouts": 0,
5
+ "expected_rollouts": 200,
6
+ "ledger_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-s-fs/native_source_retention200/step300000_20260911/episode_ledger.json",
7
+ "method": "fs",
8
+ "missing_rollouts": 0,
9
+ "observed_rollouts": 200,
10
+ "panel_type": "source_retention",
11
+ "protocol": "twoarm_native_source200",
12
+ "rollouts_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-s-fs/native_source_retention200/step300000_20260911/rollouts.jsonl",
13
+ "schema": "sealed_evaluation_result.v1",
14
+ "summary": {
15
+ "by_task_mode": {
16
+ "place_return:mode0": {
17
+ "success": 32,
18
+ "success_rate": 0.64,
19
+ "total": 50
20
+ },
21
+ "place_return:mode1": {
22
+ "success": 33,
23
+ "success_rate": 0.66,
24
+ "total": 50
25
+ },
26
+ "wipe:mode0": {
27
+ "success": 5,
28
+ "success_rate": 0.1,
29
+ "total": 50
30
+ },
31
+ "wipe:mode1": {
32
+ "success": 5,
33
+ "success_rate": 0.1,
34
+ "total": 50
35
+ }
36
+ },
37
+ "overall": {
38
+ "success": 75,
39
+ "success_rate": 0.375,
40
+ "total": 200
41
+ }
42
+ },
43
+ "task_family": "two_arm_placewipe"
44
+ }
data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-xs-fs/earlystop_full_validation_selected_horizon1800/step100000_20260911/fromscratch-shoulderonly_placewipe_resnet18_rf15/eval_20260911_173644.json ADDED
The diff for this file is too large to render. See raw diff
 
data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-xs-fs/native_source_retention200/step100000_20260911/evaluation.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "complete": true,
3
+ "contract_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-xs-fs/native_source_retention200/step100000_20260911/evaluation_contract.json",
4
+ "duplicate_rollouts": 0,
5
+ "expected_rollouts": 200,
6
+ "ledger_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-xs-fs/native_source_retention200/step100000_20260911/episode_ledger.json",
7
+ "method": "fs",
8
+ "missing_rollouts": 0,
9
+ "observed_rollouts": 200,
10
+ "panel_type": "source_retention",
11
+ "protocol": "twoarm_native_source200",
12
+ "rollouts_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260909_xs_recovery_attempt2/evaluations/capacity-xs-fs/native_source_retention200/step100000_20260911/rollouts.jsonl",
13
+ "schema": "sealed_evaluation_result.v1",
14
+ "summary": {
15
+ "by_task_mode": {
16
+ "place_return:mode0": {
17
+ "success": 34,
18
+ "success_rate": 0.68,
19
+ "total": 50
20
+ },
21
+ "place_return:mode1": {
22
+ "success": 28,
23
+ "success_rate": 0.56,
24
+ "total": 50
25
+ },
26
+ "wipe:mode0": {
27
+ "success": 3,
28
+ "success_rate": 0.06,
29
+ "total": 50
30
+ },
31
+ "wipe:mode1": {
32
+ "success": 4,
33
+ "success_rate": 0.08,
34
+ "total": 50
35
+ }
36
+ },
37
+ "overall": {
38
+ "success": 69,
39
+ "success_rate": 0.345,
40
+ "total": 200
41
+ }
42
+ },
43
+ "task_family": "two_arm_placewipe"
44
+ }
data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-m-coord/provisional_full_20260913/step200000/native_source_retention200/evaluation.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "complete": true,
3
+ "contract_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-m-coord/provisional_full_20260913/step200000/native_source_retention200/evaluation_contract.json",
4
+ "duplicate_rollouts": 0,
5
+ "expected_rollouts": 200,
6
+ "ledger_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-m-coord/provisional_full_20260913/step200000/native_source_retention200/episode_ledger.json",
7
+ "method": "coord",
8
+ "missing_rollouts": 0,
9
+ "observed_rollouts": 200,
10
+ "panel_type": "source_retention",
11
+ "protocol": "twoarm_native_source200",
12
+ "rollouts_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-m-coord/provisional_full_20260913/step200000/native_source_retention200/rollouts.jsonl",
13
+ "schema": "sealed_evaluation_result.v1",
14
+ "summary": {
15
+ "by_task_mode": {
16
+ "place_return:mode0": {
17
+ "success": 50,
18
+ "success_rate": 1.0,
19
+ "total": 50
20
+ },
21
+ "place_return:mode1": {
22
+ "success": 50,
23
+ "success_rate": 1.0,
24
+ "total": 50
25
+ },
26
+ "wipe:mode0": {
27
+ "success": 45,
28
+ "success_rate": 0.9,
29
+ "total": 50
30
+ },
31
+ "wipe:mode1": {
32
+ "success": 49,
33
+ "success_rate": 0.98,
34
+ "total": 50
35
+ }
36
+ },
37
+ "overall": {
38
+ "success": 194,
39
+ "success_rate": 0.97,
40
+ "total": 200
41
+ }
42
+ },
43
+ "task_family": "two_arm_placewipe"
44
+ }
data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-m-coord/provisional_full_20260913/step200000/twoarm_full200/coord-shoulderonly_placewipe_resnet18_rf15/eval_20260913_173448.json ADDED
The diff for this file is too large to render. See raw diff
 
data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-s-coord/provisional_full_20260913/step100000/native_source_retention200/evaluation.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "complete": true,
3
+ "contract_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-s-coord/provisional_full_20260913/step100000/native_source_retention200/evaluation_contract.json",
4
+ "duplicate_rollouts": 0,
5
+ "expected_rollouts": 200,
6
+ "ledger_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-s-coord/provisional_full_20260913/step100000/native_source_retention200/episode_ledger.json",
7
+ "method": "coord",
8
+ "missing_rollouts": 0,
9
+ "observed_rollouts": 200,
10
+ "panel_type": "source_retention",
11
+ "protocol": "twoarm_native_source200",
12
+ "rollouts_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-s-coord/provisional_full_20260913/step100000/native_source_retention200/rollouts.jsonl",
13
+ "schema": "sealed_evaluation_result.v1",
14
+ "summary": {
15
+ "by_task_mode": {
16
+ "place_return:mode0": {
17
+ "success": 50,
18
+ "success_rate": 1.0,
19
+ "total": 50
20
+ },
21
+ "place_return:mode1": {
22
+ "success": 50,
23
+ "success_rate": 1.0,
24
+ "total": 50
25
+ },
26
+ "wipe:mode0": {
27
+ "success": 46,
28
+ "success_rate": 0.92,
29
+ "total": 50
30
+ },
31
+ "wipe:mode1": {
32
+ "success": 48,
33
+ "success_rate": 0.96,
34
+ "total": 50
35
+ }
36
+ },
37
+ "overall": {
38
+ "success": 194,
39
+ "success_rate": 0.97,
40
+ "total": 200
41
+ }
42
+ },
43
+ "task_family": "two_arm_placewipe"
44
+ }
data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-s-coord/provisional_full_20260913/step100000/twoarm_full200/coord-shoulderonly_placewipe_resnet18_rf15/eval_20260913_173446.json ADDED
The diff for this file is too large to render. See raw diff
 
data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-xs-coord/provisional_full_20260913/step100000/native_source_retention200/evaluation.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "complete": true,
3
+ "contract_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-xs-coord/provisional_full_20260913/step100000/native_source_retention200/evaluation_contract.json",
4
+ "duplicate_rollouts": 0,
5
+ "expected_rollouts": 200,
6
+ "ledger_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-xs-coord/provisional_full_20260913/step100000/native_source_retention200/episode_ledger.json",
7
+ "method": "coord",
8
+ "missing_rollouts": 0,
9
+ "observed_rollouts": 200,
10
+ "panel_type": "source_retention",
11
+ "protocol": "twoarm_native_source200",
12
+ "rollouts_path": "artifact://data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-xs-coord/provisional_full_20260913/step100000/native_source_retention200/rollouts.jsonl",
13
+ "schema": "sealed_evaluation_result.v1",
14
+ "summary": {
15
+ "by_task_mode": {
16
+ "place_return:mode0": {
17
+ "success": 50,
18
+ "success_rate": 1.0,
19
+ "total": 50
20
+ },
21
+ "place_return:mode1": {
22
+ "success": 49,
23
+ "success_rate": 0.98,
24
+ "total": 50
25
+ },
26
+ "wipe:mode0": {
27
+ "success": 46,
28
+ "success_rate": 0.92,
29
+ "total": 50
30
+ },
31
+ "wipe:mode1": {
32
+ "success": 49,
33
+ "success_rate": 0.98,
34
+ "total": 50
35
+ }
36
+ },
37
+ "overall": {
38
+ "success": 194,
39
+ "success_rate": 0.97,
40
+ "total": 200
41
+ }
42
+ },
43
+ "task_family": "two_arm_placewipe"
44
+ }
data_primary/results/twoarm_forwardonly_capacity_ablation/20260912_coord_wd1e4_control/evaluations/capacity-xs-coord/provisional_full_20260913/step100000/twoarm_full200/coord-shoulderonly_placewipe_resnet18_rf15/eval_20260913_173445.json ADDED
The diff for this file is too large to render. See raw diff