auto-sync 2026-07-04T04:28:21Z workspace (part 6)

workspace/runs/ctt_base_context_obs_test_perdim_trainmax_rollout_comparison/command.txt

@@ -1 +1 @@
-python scripts/build_ctt_rollout_comparison.py --run-glob runs/ctt_residual_base_context_obs_rollout_test_perdim_trainmax_seed* --out-dir runs/ctt_base_context_obs_test_perdim_trainmax_rollout_comparison --k 8 --bootstrap-samples 1000
+python scripts/build_ctt_rollout_comparison.py --run-glob runs/ctt_residual_base_context_obs_rollout_test_perdim_trainmax_seed* --out-dir runs/ctt_base_context_obs_test_perdim_trainmax_rollout_comparison --k 8 --bootstrap-samples 1000 --no-markdown-report

workspace/runs/ctt_base_context_obs_test_perdim_trainmax_rollout_comparison/metrics.json

@@ -1,4 +1,5 @@
 {
+  "data_hash": "0c38d73dc9bc4ab756e12006671b5d12f1a23358b281e54b3f6a9ae3cbf6e9af",
   "k": 8,
   "num_rows": 144,
   "report_type": "ctt_measured_rollout_comparison",
@@ -8,6 +9,7 @@
     "runs/ctt_residual_base_context_obs_rollout_test_perdim_trainmax_seed2"
   ],
   "source_content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
+  "split_hash": "84151df9e88c219b76751677558cc5bbbc19082bf358cea7330158e9e403e25f",
   "success_summary": {
     "any_generated_unsafe_known": 0.3680555555555556,
     "base_safety_label_known_rate": 1.0,

workspace/runs/ctt_base_context_obs_train_cal_tanh_rollout_comparison/command.txt

@@ -1 +1 @@
-python scripts/build_ctt_rollout_comparison.py --run-glob runs/ctt_residual_base_context_obs_rollout_train_cal_tanh_seed* --out-dir runs/ctt_base_context_obs_train_cal_tanh_rollout_comparison --k 8
+python scripts/build_ctt_rollout_comparison.py --run-glob runs/ctt_residual_base_context_obs_rollout_train_cal_tanh_seed* --out-dir runs/ctt_base_context_obs_train_cal_tanh_rollout_comparison --k 8 --bootstrap-samples 1000 --no-markdown-report

workspace/runs/ctt_base_context_obs_train_cal_tanh_rollout_comparison/metrics.json

@@ -1,4 +1,5 @@
 {
+  "data_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
   "k": 8,
   "num_rows": 432,
   "report_type": "ctt_measured_rollout_comparison",
@@ -8,6 +9,7 @@
     "runs/ctt_residual_base_context_obs_rollout_train_cal_tanh_seed2"
   ],
   "source_content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
+  "split_hash": "867a94a1c67686c082765671a7bf8d74a9d633085e1b2619c83197c60bf58372",
   "success_summary": {
     "any_generated_unsafe_known": 0.0,
     "base_safety_label_known_rate": 1.0,

workspace/runs/ctt_base_context_obs_val_perdim_trainmax_rollout_comparison/command.txt

@@ -1 +1 @@
-python scripts/build_ctt_rollout_comparison.py --run-glob runs/ctt_residual_base_context_obs_rollout_val69_perdim_trainmax_seed* --out-dir runs/ctt_base_context_obs_val_perdim_trainmax_rollout_comparison --k 8 --bootstrap-samples 1000
+python scripts/build_ctt_rollout_comparison.py --run-glob runs/ctt_residual_base_context_obs_rollout_val69_perdim_trainmax_seed* --out-dir runs/ctt_base_context_obs_val_perdim_trainmax_rollout_comparison --k 8 --bootstrap-samples 1000 --no-markdown-report

workspace/runs/ctt_base_context_obs_val_perdim_trainmax_rollout_comparison/metrics.json

@@ -1,4 +1,5 @@
 {
+  "data_hash": "1aacf9de98648fa36bef14e137eb43c7c6973170804650814dae41d09842c708",
   "k": 8,
   "num_rows": 207,
   "report_type": "ctt_measured_rollout_comparison",
@@ -8,6 +9,7 @@
     "runs/ctt_residual_base_context_obs_rollout_val69_perdim_trainmax_seed2"
   ],
   "source_content_hash": "67f7b4a692f7c7e71da24378dc71e3399c5d35afc14c494869eba6087b765c42",
+  "split_hash": "9d1dc1c10868f76fa080895df0f2730c173f12b6db14932599d55c2b60859381",
   "success_summary": {
     "any_generated_unsafe_known": 0.3285024154589372,
     "base_safety_label_known_rate": 1.0,