[07:59:27] longeval: T=16384 prompt_len=4096 max_new=24 n=100 greedy=1 gpus=0,1
[07:59:27] haystack: /root/Adhi/BluTrain/dist/Context_Parallelism/lendata/haystack/b2val_haystack.bin
[07:59:27] prompt signature changed -- regenerating prompts and clearing state
[07:59:27] generating prompts: niah_single
[07:59:27] fan-out: 2 GPU(s), NP=1 -> 4 concurrent job(s)
[07:59:27] 4 job(s) to run
[07:59:27]   [gpu 0] armA_c / niah_single
[07:59:27]   [gpu 1] l1 / niah_single
[07:59:27]   [gpu 0] l2 / niah_single
[07:59:27]   [gpu 1] l3 / niah_single
[08:04:42]   scoring armA_c / niah_single

=== niah_single  (prompt_len=4096, n=100) ===
  accuracy: 73/100 = 73.0%
    depth      acc   (needle position in context)
     0.00  11/20   55.0%
     0.25  14/20   70.0%
     0.50  10/20   50.0%
     0.75  19/20   95.0%
     1.00  19/20   95.0%

  --- 2 sample generations ---
  [HIT ] expect=8056020
         got= 8056020.\n
  [HIT ] expect=5343902
         got= 5343902.\n

  wrote decide_len_4096/scores/armA_c_niah_single.csv
[08:04:42]   scoring l1 / niah_single

=== niah_single  (prompt_len=4096, n=100) ===
  accuracy: 75/100 = 75.0%
    depth      acc   (needle position in context)
     0.00  10/20   50.0%
     0.25  16/20   80.0%
     0.50  12/20   60.0%
     0.75  19/20   95.0%
     1.00  18/20   90.0%

  --- 2 sample generations ---
  [HIT ] expect=8056020
         got= 8056020.\n
  [HIT ] expect=5343902
         got= 5343902.\n

  wrote decide_len_4096/scores/l1_niah_single.csv
[08:04:42]   scoring l2 / niah_single

=== niah_single  (prompt_len=4096, n=100) ===
  accuracy: 80/100 = 80.0%
    depth      acc   (needle position in context)
     0.00  15/20   75.0%
     0.25  17/20   85.0%
     0.50  12/20   60.0%
     0.75  18/20   90.0%
     1.00  18/20   90.0%

  --- 2 sample generations ---
  [HIT ] expect=8056020
         got= 80560202020202020202020202020202020202020202020
  [HIT ] expect=5343902
         got= 5343902.\n

  wrote decide_len_4096/scores/l2_niah_single.csv
[08:04:43]   scoring l3 / niah_single

=== niah_single  (prompt_len=4096, n=100) ===
  accuracy: 74/100 = 74.0%
    depth      acc   (needle position in context)
     0.00  13/20   65.0%
     0.25  17/20   85.0%
     0.50  10/20   50.0%
     0.75  16/20   80.0%
     1.00  18/20   90.0%

  --- 2 sample generations ---
  [HIT ] expect=8056020
         got= 80560202020202020202020202020202020202020202020
  [HIT ] expect=5343902
         got= 5343902.\n

  wrote decide_len_4096/scores/l3_niah_single.csv
[08:04:43] building summary

=== per-task accuracy %  (secondary/descriptive: ~10 per depth cell) ===
  task                 armA_c       l1       l2       l3
  niah_single            73.0     75.0     80.0     74.0

=== PRIMARY: accuracy % pooled by depth across tasks (+-95% Wilson) ===
  depth             armA_c              l1              l2              l3
  0.00          55.0+-20.0      50.0+-20.1      75.0+-17.8      65.0+-19.3
  0.25          70.0+-18.7      80.0+-16.8      85.0+-15.4      85.0+-15.4
  0.50          50.0+-20.1      60.0+-19.7      60.0+-19.7      50.0+-20.1
  0.75          95.0+-11.4      95.0+-11.4      90.0+-13.7      80.0+-16.8
  1.00          95.0+-11.4      90.0+-13.7      90.0+-13.7      90.0+-13.7

  wrote decide_len_4096/summary.csv and decide_len_4096/depth_summary.csv
[08:04:43] DONE -> decide_len_4096/summary.csv