fo2_sim4: telemetry run complete
Browse files- runs/fo2_sim4/eval/val_step_000_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_000_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_008_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_008_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_016_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_016_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_024_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_024_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_032_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_032_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_040_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_040_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_048_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_048_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_056_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_056_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_064_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_064_natural.parquet +3 -0
- runs/fo2_sim4/eval/val_step_071_forced.parquet +3 -0
- runs/fo2_sim4/eval/val_step_071_natural.parquet +3 -0
- runs/fo2_sim4/metrics.csv +73 -0
- runs/fo2_sim4/run_config.json +39 -0
- runs/fo2_sim4/state.json +5 -0
- runs/fo2_sim4/wandb_run_id.txt +1 -0
runs/fo2_sim4/eval/val_step_000_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4f8da1240b0c0bf125f61701528c7c7af39bc4c4d2c15c4e8d08748a8e3ef5c1
|
| 3 |
+
size 268269
|
runs/fo2_sim4/eval/val_step_000_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b569f85e2ed9c7a79952c60b48f63c11b2c082fd61200911a3878df47f5930c
|
| 3 |
+
size 273507
|
runs/fo2_sim4/eval/val_step_008_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1be7d771f4abb2b6afa5e0be6a4dd95c43aa9d63f293dff4d89d8c3e0a3d4c24
|
| 3 |
+
size 301714
|
runs/fo2_sim4/eval/val_step_008_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d85bf6f89e35245571699ca91e3363697d0a431c9cecbab9f3fb8309e4d072f4
|
| 3 |
+
size 330534
|
runs/fo2_sim4/eval/val_step_016_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04fbae4c4abb93f14bc8c791253081ce430b98fc544e0f0d17fd83104bccce0f
|
| 3 |
+
size 287409
|
runs/fo2_sim4/eval/val_step_016_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36e09789a517da545bf1311b7dfac05753418e0e2b5cebcbe61c80359a771c1b
|
| 3 |
+
size 262075
|
runs/fo2_sim4/eval/val_step_024_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af8c23c706fee545c6b28fc3387a8601f55798ac44b342dca09d839cb0678fe2
|
| 3 |
+
size 293885
|
runs/fo2_sim4/eval/val_step_024_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87d9b671f188de32cb5a1f811ea46dc0f9aa76e1d6ff477da0aad285ff10f3df
|
| 3 |
+
size 278234
|
runs/fo2_sim4/eval/val_step_032_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ef6593776e902276776f5b106b2fa76cbee229a55dce4d00f576da400fc1fb3
|
| 3 |
+
size 274955
|
runs/fo2_sim4/eval/val_step_032_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69f83c7fe86d91e4e9f416a69358ce3d011295a84ae5fddd498043072b9bed84
|
| 3 |
+
size 267904
|
runs/fo2_sim4/eval/val_step_040_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c3711584b78e2090fca0693abe5e8a831258a37b5789c12b82324f9416b95c1
|
| 3 |
+
size 274974
|
runs/fo2_sim4/eval/val_step_040_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05e00e82df7f554ed95c7e37d7398f0cd3287b55ed9e857a6c48a071d1202948
|
| 3 |
+
size 254095
|
runs/fo2_sim4/eval/val_step_048_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e0c90006a28075ae781ca31fe874389d785527d66fe131ef8471f3f9e4c5550
|
| 3 |
+
size 290423
|
runs/fo2_sim4/eval/val_step_048_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e75e7a51a01c08150a5d8cf2ef164de3b4c0e9e97556524225ed898dcb45aa8
|
| 3 |
+
size 241284
|
runs/fo2_sim4/eval/val_step_056_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83e9f0a39264d85bc1e333e1dc87d347c2f4c131315534aee9788bd542173dc2
|
| 3 |
+
size 273561
|
runs/fo2_sim4/eval/val_step_056_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6cb71c1e0d59637ab2e99b1e893d800545483611824c3b78e0251c2356b199b
|
| 3 |
+
size 260320
|
runs/fo2_sim4/eval/val_step_064_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3798354715b433dd0bd843b9ebbd24b433fd3f3ee864377ec5c3080c336a0dc7
|
| 3 |
+
size 286613
|
runs/fo2_sim4/eval/val_step_064_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0cfe625cd74d42480dcca5f32850e55d74960d0cebce50a8696d496f397501d4
|
| 3 |
+
size 270612
|
runs/fo2_sim4/eval/val_step_071_forced.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f38057d1c22a6f259f148fca6fc1abcd71f4f9bb2039327deb990d5b3b48869
|
| 3 |
+
size 265056
|
runs/fo2_sim4/eval/val_step_071_natural.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:243ac561f1139d1a206bd9f81e373b086e31c1cac044804f5679203d4acf7d0d
|
| 3 |
+
size 260246
|
runs/fo2_sim4/metrics.csv
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
step,train/loss,train/grad_norm,train/n_loss_tokens,train/n_comp_tokens,batch/n_seqs,batch/n_real_tokens,batch/n_packs,time/train_s,train/epochs,train/lr,holdout/ce_branch_correct,holdout/ce_branch_incorrect,holdout/ce_branch,holdout/val_branch_frac,holdout/n_val_rollouts,time/holdout_ce_s,time/gen_setup_s,time/gen_grade_s,time/gen_decode_forced_s,val/forced_rate,val/forced_cov50,val/forced_pass_at_k,val/forced_unresolved,length/forced_mean_toks,length/forced_p90_toks,length/forced_trunc_frac,length/forced_correct_mean_toks,length/forced_incorrect_mean_toks,time/gen_decode_natural_s,val/natural_rate,val/natural_cov50,val/natural_pass_at_k,val/natural_unresolved,length/natural_mean_toks,length/natural_p90_toks,length/natural_trunc_frac,length/natural_correct_mean_toks,length/natural_incorrect_mean_toks,time/gen_teardown_s,time/gen_val_s,time/wall_step,time/wall_total,time/train_frac,time/val_frac
|
| 2 |
+
0,0.1427204645178318,1.662854552268982,30311,30311,86,36484,9,40.99929705900104,0.08022388059701492,5e-06,0.1380109186672716,,0.1380109186672716,1.0,114,7.204624944000898,1.415128231048584,13.852715015411377,96.62107539176941,0.325,0.25,0.85,0.0,530.83125,1069.6000000000001,0.0625,505.37735849056605,543.4392523364486,90.4953203201294,0.184375,0.025,0.7,0.0,508.15625,1099.3000000000002,0.053125,501.4,509.7153846153846,0.6082100868225098,204.2967868529995,254.54175758361816,254.54175758361816,0.1610710063771465,0.8309104714479753
|
| 3 |
+
1,0.13228315317690484,1.440353274345398,30344,30344,72,36235,9,7.7747284099987155,0.14738805970149255,5e-06,0.13539098100329433,,0.13539098100329433,1.0,114,7.004502870000579,,,,,,,,,,,,,,,,,,,,,,,,,16.451719284057617,270.9934768676758,0.17998228604157793,0.8063142965382001
|
| 4 |
+
2,0.1115670684790972,1.0152946710586548,31192,31192,61,36042,9,7.165490032999514,0.2042910447761194,5e-06,0.1329491991004675,,0.1329491991004675,1.0,114,7.005750169999374,,,,,,,,,,,,,,,,,,,,,,,,,15.646647930145264,286.64012479782104,0.19515591385349726,0.7867414410179416
|
| 5 |
+
3,0.11993360733480872,1.0269211530685425,31630,31630,69,37148,10,8.012587238001288,0.26865671641791045,5e-06,0.1303132335825459,,0.1303132335825459,1.0,114,7.0439920789995085,,,,,,,,,,,,,,,,,,,,,,,,,17.221440076828003,303.86156487464905,0.21046460010953505,0.7653342304477871
|
| 6 |
+
4,0.1302388832454042,1.2860051393508911,30517,30517,78,37118,10,8.000292552000246,0.3414179104477612,5e-06,0.12855386838419505,,0.12855386838419505,1.0,114,7.005818399000418,,,,,,,,,,,,,,,,,,,,,,,,,16.614584922790527,320.4761497974396,0.22451716091034885,0.747517328407801
|
| 7 |
+
5,0.12387842790345002,1.1937850713729858,31652,31652,74,37109,10,7.994591028000286,0.41044776119402987,5e-06,0.12630627009646303,,0.12630627009646303,1.0,114,7.008395524999287,,,,,,,,,,,,,,,,,,,,,,,,,16.64153552055359,337.11768531799316,0.23714859766134622,0.7314059201830883
|
| 8 |
+
6,0.1335586640787096,1.1088862419128418,29882,29882,83,37135,10,7.98145278300035,0.4878731343283582,5e-06,0.12455122008955871,,0.12455122008955871,1.0,114,7.022526188000484,,,,,,,,,,,,,,,,,,,,,,,,,17.18924641609192,354.3069317340851,0.2481702479617125,0.715742127276038
|
| 9 |
+
7,0.13077600835400077,0.984188973903656,30644,30644,85,37366,10,7.974888679000287,0.5671641791044776,5e-06,0.12336392598583631,,0.12336392598583631,1.0,114,7.0072166720001405,,,,,,,,,,,,,,,,,,,,,,,,,16.59942102432251,370.9063527584076,0.2585648023248311,0.7026021845189142
|
| 10 |
+
8,0.10921860129303272,0.9327333569526672,30471,30471,71,36526,9,7.162899923001532,0.6333955223880597,5e-06,0.12254157378730791,,0.12254157378730791,1.0,114,7.006895604999954,1.4549949169158936,13.177685976028442,95.33488774299622,0.3625,0.35,0.825,0.0,545.00625,1087.5000000000005,0.078125,428.57758620689657,611.2107843137255,94.777996301651,0.240625,0.2,0.7,0.0,554.453125,1467.4000000000015,0.090625,490.75949367088606,575.3319502074689,0.7682898044586182,206.10533827899962,222.17994332313538,593.086296081543,0.17377947928648946,0.7987233067325324
|
| 11 |
+
9,0.11916620848022251,1.070743441581726,30919,30919,78,37372,10,8.12311269699967,0.7061567164179104,5e-06,0.12194915964728859,,0.12194915964728859,1.0,114,7.037828767000974,,,,,,,,,,,,,,,,,,,,,,,,,17.416536569595337,610.5028326511383,0.182127476655854,0.7874651035824385
|
| 12 |
+
10,0.1258159268929504,0.9868577718734741,30640,30640,79,37417,10,7.9831845270000485,0.7798507462686567,5e-06,0.12125009863294735,,0.12125009863294735,1.0,114,7.0083881489990745,,,,,,,,,,,,,,,,,,,,,,,,,16.657716751098633,627.1605494022369,0.19001916661146723,0.777724404006111
|
| 13 |
+
11,0.11696623970133882,0.8307859897613525,31072,31072,77,36902,10,8.008411630999035,0.851679104477612,5e-06,0.12093817193695382,,0.12093817193695382,1.0,114,7.007703855000727,,,,,,,,,,,,,,,,,,,,,,,,,16.6344895362854,643.7950389385223,0.1975487987134782,0.7685144159712095
|
| 14 |
+
12,0.12681009515928837,1.0326249599456787,29004,29004,83,36121,9,7.133550759999707,0.9291044776119403,5e-06,0.12084200481328783,,0.12084200481328783,1.0,114,7.034245026001372,,,,,,,,,,,,,,,,,,,,,,,,,16.44126868247986,660.2363076210022,0.20343396109791456,0.760030927697863
|
| 15 |
+
13,0.10733927250506545,0.9239827394485474,31093,31093,69,37054,10,8.028544494000016,0.9934701492537313,5e-06,0.12032418183970173,,0.12032418183970173,1.0,114,7.007508595999752,,,,,,,,,,,,,,,,,,,,,,,,,16.527619123458862,676.7639267444611,0.21032892887588697,0.7518242357044571
|
| 16 |
+
14,0.0997352162400706,1.9516253471374512,3399,3399,7,4169,2,1.6744881100003113,1.0,5e-06,0.11993396524174936,,0.11993396524174936,1.0,114,7.010745850000603,,,,,,,,,,,,,,,,,,,,,,,,,10.324212551116943,687.088139295578,0.20960559742983315,0.7507308572606618
|
| 17 |
+
15,0.11436282639092121,0.9130328297615051,30753,30753,82,37270,10,7.989298358001179,1.0764925373134329,5e-06,0.11953080306945732,,0.11953080306945732,1.0,114,7.023752577999403,,,,,,,,,,,,,,,,,,,,,,,,,17.293102741241455,704.3812420368195,0.21580191125256784,0.7422713570468301
|
| 18 |
+
16,0.10729159992311142,0.9035354852676392,31214,31214,74,37195,10,8.010046113999124,1.1455223880597014,5e-06,0.11926326119977117,,0.11926326119977117,1.0,114,7.007591609999508,1.463285207748413,13.040749073028564,95.07684421539307,0.353125,0.375,0.825,0.0,496.953125,901.7000000000006,0.053125,461.1913043478261,517.0146341463415,88.5384783744812,0.23125,0.225,0.675,0.0,463.984375,984.6000000000013,0.053125,412.26666666666665,479.81632653061223,0.6681907176971436,199.40716019700085,215.95421385765076,920.3354558944702,0.17386797756311867,0.7923814817101015
|
| 19 |
+
17,0.1024330900243309,0.7177847027778625,30825,30825,79,37385,10,8.411689028000183,1.2192164179104477,5e-06,0.11904565225968082,,0.11904565225968082,1.0,114,7.00627311900098,,,,,,,,,,,,,,,,,,,,,,,,,16.939563751220703,937.2750196456909,0.1797002479460852,0.7855357604743648
|
| 20 |
+
18,0.11128544122331953,0.7223068475723267,31390,31390,77,37151,10,8.005081014000098,1.291044776119403,5e-06,0.11910174974848599,,0.11910174974848599,1.0,114,7.023010168999463,,,,,,,,,,,,,,,,,,,,,,,,,17.311251163482666,954.5862708091736,0.1848273328805004,0.7786473346929048
|
| 21 |
+
19,0.1138401403956605,0.9549148678779602,31340,31340,79,37132,10,7.988362056001279,1.3647388059701493,5e-06,0.11912209279387687,,0.11912209279387687,1.0,114,7.006492861000879,,,,,,,,,,,,,,,,,,,,,,,,,16.502323627471924,971.0885944366455,0.18991263778666048,0.7726303785868973
|
| 22 |
+
20,0.09265607799968549,0.7437169551849365,31795,31795,72,37241,10,7.984614271999817,1.4319029850746268,5e-06,0.11915784723729114,,0.11915784723729114,1.0,114,7.007305248000193,,,,,,,,,,,,,,,,,,,,,,,,,16.618993282318115,987.7075877189636,0.1948011872727963,0.7667247503463354
|
| 23 |
+
21,0.09769062170031677,0.7682950496673584,31253,31253,74,36767,10,8.012212712999826,1.5009328358208955,5e-06,0.11921209535833349,,0.11921209535833349,1.0,114,7.042396284999995,,,,,,,,,,,,,,,,,,,,,,,,,17.08236527442932,1004.789952993393,0.19946340315399375,0.760698539646952
|
| 24 |
+
22,0.09446953313827758,0.6877001523971558,31444,31444,71,37413,10,8.016524089000086,1.5671641791044777,5e-06,0.11947223975696841,,0.11947223975696841,1.0,114,7.0047327620013675,,,,,,,,,,,,,,,,,,,,,,,,,16.546232223510742,1021.3361852169037,0.20408103676825834,0.7552331874858542
|
| 25 |
+
23,0.10610316649642493,0.7394396066665649,31328,31328,77,37279,10,7.99658249300046,1.6389925373134329,5e-06,0.11936189414712091,,0.11936189414712091,1.0,114,7.007350787000178,,,,,,,,,,,,,,,,,,,,,,,,,16.593008279800415,1037.929193496704,0.20852282739235947,0.7499108208150392
|
| 26 |
+
24,0.09781375901411789,0.775284469127655,29537,29537,74,36812,10,8.014643546001025,1.7080223880597014,5e-06,0.1192928510839761,,0.1192928510839761,1.0,114,7.023200600000564,1.458425760269165,13.531769752502441,91.67919421195984,0.409375,0.475,0.775,0.0,498.015625,964.5000000000003,0.04375,433.41984732824426,542.7883597883598,88.32923412322998,0.2125,0.175,0.7,0.0,481.396875,1050.0000000000007,0.05625,440.65714285714284,492.804,0.5194158554077148,196.1331283090003,213.56395196914673,1251.4931454658508,0.17934303069910784,0.7842717044899613
|
| 27 |
+
25,0.10293550563964138,0.7809187173843384,31119,31119,71,36855,10,8.127374376001171,1.7742537313432836,5e-06,0.11937483972146055,,0.11937483972146055,1.0,114,7.004456225000467,,,,,,,,,,,,,,,,,,,,,,,,,16.738999366760254,1268.232144832611,0.1833843661277824,0.7794433555439305
|
| 28 |
+
26,0.10348000260552372,0.8095805644989014,30704,30704,76,37270,10,7.9876880689989775,1.8451492537313432,5e-06,0.11923182194780345,,0.11923182194780345,1.0,114,7.006109191001087,,,,,,,,,,,,,,,,,,,,,,,,,16.60919690132141,1284.8413417339325,0.18723061613767816,0.7748203575271958
|
| 29 |
+
27,0.10361554988586008,0.8126201629638672,29788,29788,84,37221,10,7.968395625999619,1.9235074626865671,5e-06,0.11930641311423668,,0.11930641311423668,1.0,114,7.038712465999197,,,,,,,,,,,,,,,,,,,,,,,,,17.264994382858276,1302.1063361167908,0.19086769243377164,0.7699524320140344
|
| 30 |
+
28,0.10549455085865257,0.8390846848487854,30280,30280,82,37207,10,7.987698792998344,2.0,5e-06,0.1192638776556921,,0.1192638776556921,1.0,114,7.006054198998754,,,,,,,,,,,,,,,,,,,,,,,,,16.585434436798096,1318.6917705535889,0.19452440380614242,0.765581477777167
|
| 31 |
+
29,0.09260446039477058,0.7670801281929016,31208,31208,72,37284,10,7.991770030001135,2.0671641791044775,5e-06,0.11917325863531454,,0.11917325863531454,1.0,114,7.008112034000078,,,,,,,,,,,,,,,,,,,,,,,,,16.612242460250854,1335.3040130138397,0.1980893473868882,0.7613053630929733
|
| 32 |
+
30,0.10306479573479868,0.7975302338600159,30573,30573,82,37222,10,7.974369829000352,2.1436567164179103,5e-06,0.11922134219714753,,0.11922134219714753,1.0,114,7.042047909999383,,,,,,,,,,,,,,,,,,,,,,,,,17.160939693450928,1352.4649527072906,0.20147203798853447,0.7568522587797827
|
| 33 |
+
31,0.09233076947627103,0.7404676079750061,31333,31333,78,37302,10,7.980645352001375,2.216417910447761,5e-06,0.11904811808336457,,0.11904811808336457,1.0,114,7.007596937999551,,,,,,,,,,,,,,,,,,,,,,,,,16.5538432598114,1369.018795967102,0.2048653506498283,0.7528192851340302
|
| 34 |
+
32,0.09419996764277624,0.6500082015991211,30905,30905,72,36987,10,8.001777019999281,2.283582089552239,5e-06,0.11956532460102973,,0.11956532460102973,1.0,114,7.011019158000636,1.4611907005310059,13.742653608322144,90.2486834526062,0.421875,0.475,0.775,0.0,486.646875,932.8000000000002,0.04375,409.3925925925926,543.0216216216216,86.92660665512085,0.19375,0.15,0.675,0.0,463.428125,1040.6000000000001,0.05625,414.7096774193548,475.1356589147287,0.6206669807434082,193.68215973900078,210.6723074913025,1579.6911034584045,0.18260930385090385,0.7794669017988985
|
| 35 |
+
33,0.09521583677753723,0.7141545414924622,30486,30486,79,37290,10,8.116982775998622,2.357276119402985,5e-06,0.11983040064703214,,0.11983040064703214,1.0,114,7.043028609999965,,,,,,,,,,,,,,,,,,,,,,,,,17.374157428741455,1597.065260887146,0.18570516981457377,0.7753972171018956
|
| 36 |
+
34,0.0965281173594132,1.0296306610107422,30675,30675,85,37027,10,7.980548505000115,2.4365671641791047,5e-06,0.11978293354112007,,0.11978293354112007,1.0,114,7.008640171001389,,,,,,,,,,,,,,,,,,,,,,,,,16.557142972946167,1613.6224038600922,0.18874541110387996,0.7717844001259826
|
| 37 |
+
35,0.08495861478451146,0.6589687466621399,31533,31533,64,37440,10,8.01659168399965,2.496268656716418,5e-06,0.11965717653324917,,0.11965717653324917,1.0,114,7.008296111000163,,,,,,,,,,,,,,,,,,,,,,,,,16.734060525894165,1630.3564643859863,0.19172519782949776,0.7681613944320262
|
| 38 |
+
36,0.08817700380129309,0.6960247755050659,30779,30779,72,36355,9,7.160168625001461,2.5634328358208953,5e-06,0.12006527035290869,,0.12006527035290869,1.0,114,7.045257671999934,,,,,,,,,,,,,,,,,,,,,,,,,16.39124059677124,1646.7477049827576,0.19416488835814294,0.7647936286567919
|
| 39 |
+
37,0.09618897637795276,0.7139504551887512,31750,31750,74,37178,10,7.9953540959995735,2.6324626865671643,5e-06,0.12010780581145326,,0.12010780581145326,1.0,114,7.012455320000299,,,,,,,,,,,,,,,,,,,,,,,,,16.62455654144287,1663.3722615242004,0.19703102304212408,0.7613657131299844
|
| 40 |
+
38,0.09699146834306241,0.7478383779525757,31178,31178,82,37239,10,7.979160836999654,2.708955223880597,5e-06,0.11998574753910797,,0.11998574753910797,1.0,114,7.010427864001031,,,,,,,,,,,,,,,,,,,,,,,,,16.610104084014893,1679.9823656082153,0.1998325137796689,0.7580109541798522
|
| 41 |
+
39,0.09328773399659641,0.7343448996543884,30556,30556,74,37434,10,7.9889885839984345,2.7779850746268657,5e-06,0.12003814629238752,,0.12003814629238752,1.0,114,7.02669291300117,,,,,,,,,,,,,,,,,,,,,,,,,17.21181011199951,1697.1941757202148,0.20251312002243346,0.7544638952874286
|
| 42 |
+
40,0.10328227571115974,0.8403698801994324,29705,29705,84,36459,9,7.130885462000151,2.8563432835820897,5e-06,0.12011397037066261,,0.12011397037066261,1.0,114,7.0098377210015315,1.466261386871338,17.97399878501892,91.53112769126892,0.35625,0.35,0.775,0.0,475.096875,859.6000000000001,0.034375,443.39130434782606,492.8829268292683,88.07577300071716,0.225,0.25,0.65,0.0,463.178125,978.5000000000006,0.053125,444.8904109589041,468.582995951417,0.5141379833221436,200.12155478700151,215.91955137252808,1913.113727092743,0.18338427470495874,0.7775821689605689
|
| 43 |
+
41,0.08241891237695659,0.6720669269561768,30985,30985,67,37153,10,8.427960905999498,2.9188432835820897,5e-06,0.12020890458248673,,0.12020890458248673,1.0,114,7.006995902000199,,,,,,,,,,,,,,,,,,,,,,,,,17.079598903656006,1930.193325996399,0.1861279537849116,0.7743318232185212
|
| 44 |
+
42,0.09884947583569592,0.8154100775718689,30334,30334,86,36970,10,7.971422908000022,2.9990671641791047,5e-06,0.12036116919495789,,0.12036116919495789,1.0,114,7.022980581999946,,,,,,,,,,,,,,,,,,,,,,,,,17.259368658065796,1947.4526946544647,0.18857164443224678,0.7710755193113673
|
| 45 |
+
43,0.08051948051948052,3.5782201290130615,770,770,1,858,1,0.853667615001541,3.0,5e-06,0.12043945909691674,,0.12043945909691674,1.0,114,7.008283707998999,,,,,,,,,,,,,,,,,,,,,,,,,9.454448699951172,1956.907143354416,0.18809682715402076,0.7709315113373173
|
| 46 |
+
44,0.08474466918287686,0.6610239148139954,30999,30999,74,37027,10,8.009225828000126,3.0690298507462686,5e-06,0.12056336673702484,,0.12056336673702484,1.0,114,7.007074094000927,,,,,,,,,,,,,,,,,,,,,,,,,16.61337399482727,1973.5205173492432,0.1905717458834244,0.7679922465172917
|
| 47 |
+
45,0.08576222279859941,0.7003093957901001,30844,30844,76,37088,10,8.007310489998417,3.139925373134328,5e-06,0.12081426429684572,,0.12081426429684572,1.0,114,7.033490698000605,,,,,,,,,,,,,,,,,,,,,,,,,17.636922359466553,1991.1574397087097,0.19290516829959667,0.7647220234818647
|
| 48 |
+
46,0.08310958081940803,0.6673772931098938,31657,31657,72,37254,10,8.00947008500043,3.207089552238806,5e-06,0.12098748841062869,,0.12098748841062869,1.0,114,7.008570248000979,,,,,,,,,,,,,,,,,,,,,,,,,16.564565181732178,2007.722004890442,0.19530295038251577,0.7619035468491003
|
| 49 |
+
47,0.07890149283048517,0.643399178981781,30546,30546,75,36894,10,8.011650101001578,3.27705223880597,5e-06,0.12129818219478035,,0.12129818219478035,1.0,114,7.008561525000914,,,,,,,,,,,,,,,,,,,,,,,,,16.66525101661682,2024.3872559070587,0.19765273666708696,0.7590934361269099
|
| 50 |
+
48,0.08504520622344335,0.7026116251945496,31301,31301,76,37401,10,7.979893966999953,3.34794776119403,5e-06,0.12114221884678358,,0.12114221884678358,1.0,114,7.022848890001114,1.4578499794006348,12.428373336791992,92.97879004478455,0.365625,0.375,0.75,0.0,507.1,1020.3000000000001,0.05,440.35042735042737,545.5714285714286,86.16770339012146,0.215625,0.15,0.65,0.0,431.1375,867.3000000000001,0.040625,386.1714285714286,443.728,0.6008017063140869,194.19948311100052,211.26849460601807,2235.655750513077,0.18254401424608663,0.7773653925655455
|
| 51 |
+
49,0.09012763740557436,0.6745264530181885,30712,30712,79,36647,9,7.556086933000188,3.421641791044776,5e-06,0.12118968595269564,,0.12118968595269564,1.0,114,7.008682915999088,,,,,,,,,,,,,,,,,,,,,,,,,16.12355899810791,2251.7793095111847,0.184592539929784,0.7749116823681114
|
| 52 |
+
50,0.08464001045341696,0.6480945944786072,30612,30612,80,37258,10,7.977033140999993,3.496268656716418,5e-06,0.12132962144674807,,0.12132962144674807,1.0,114,7.006669039001281,,,,,,,,,,,,,,,,,,,,,,,,,16.482994079589844,2268.2623035907745,0.18676794768151955,0.7723695620742859
|
| 53 |
+
51,0.09120449904379113,0.7777430415153503,30851,30851,82,37202,10,7.977125130000786,3.5727611940298507,5e-06,0.12189244570256248,,0.12189244570256248,1.0,114,7.042550440000923,,,,,,,,,,,,,,,,,,,,,,,,,17.293165683746338,2285.555469274521,0.18884504278166073,0.7696069232099416
|
| 54 |
+
52,0.0873248615184099,0.648049533367157,30690,30690,76,37198,10,8.000227508000535,3.6436567164179103,5e-06,0.12204840905055925,,0.12204840905055925,1.0,114,7.0072595220008225,,,,,,,,,,,,,,,,,,,,,,,,,16.612502336502075,2302.167971611023,0.19095741635888078,0.7670971857106534
|
| 55 |
+
53,0.09070781893004115,0.7201539874076843,30375,30375,78,37245,10,7.9827414190003765,3.716417910447761,5e-06,0.12214642554198804,,0.12214642554198804,1.0,114,7.010941295000521,,,,,,,,,,,,,,,,,,,,,,,,,16.4966721534729,2318.664643764496,0.19304162441330977,0.7646631944464588
|
| 56 |
+
54,0.08544508895297968,0.7149794697761536,30859,30859,75,37037,10,8.015014063999843,3.7863805970149254,5e-06,0.12226540153472866,,0.12226540153472866,1.0,114,7.046482777001074,,,,,,,,,,,,,,,,,,,,,,,,,17.18906855583191,2335.8537123203278,0.19505237034446857,0.762052857479593
|
| 57 |
+
55,0.08936655215590586,0.7226528525352478,31147,31147,76,37238,10,7.997279747998618,3.857276119402985,5e-06,0.12246266742942813,,0.12246266742942813,1.0,114,7.009382041998833,,,,,,,,,,,,,,,,,,,,,,,,,16.753159999847412,2352.606872320175,0.19706270884807242,0.7596056099281907
|
| 58 |
+
56,0.08090831027573704,0.7088336944580078,31443,31443,70,37144,10,8.00378864600134,3.922574626865672,5e-06,0.12242383070640916,,0.12242383070640916,1.0,114,7.010944913999992,1.4587748050689697,11.627954721450806,94.63667702674866,0.390625,0.425,0.75,0.0,505.825,971.6000000000001,0.059375,394.52,577.1743589743589,86.82129693031311,0.215625,0.125,0.6,0.0,452.515625,913.3000000000005,0.053125,407.5,465.12,0.5973167419433594,195.71811430600064,212.3137264251709,2564.920598745346,0.1838711389314355,0.775767654703752
|
| 59 |
+
57,0.08571475495992642,0.6786532998085022,30444,30444,82,37187,10,8.387975031999304,3.9990671641791047,5e-06,0.12198799637030754,,0.12198799637030754,1.0,114,7.035534826998628,,,,,,,,,,,,,,,,,,,,,,,,,17.61037826538086,2582.530977010727,0.18586528140994757,0.7732019441363412
|
| 60 |
+
58,0.09051724137931035,6.892394542694092,290,290,1,378,1,0.8712228040003538,4.0,5e-06,0.12219389264790011,,0.12219389264790011,1.0,114,7.006320753000182,,,,,,,,,,,,,,,,,,,,,,,,,9.446362018585205,2591.977339029312,0.1855240253667078,0.7730871187787642
|
| 61 |
+
59,0.08343554770549177,0.7012227773666382,30573,30573,81,37060,10,7.9974655020014325,4.075559701492537,5e-06,0.12235355473142248,,0.12235355473142248,1.0,114,7.010027370000898,,,,,,,,,,,,,,,,,,,,,,,,,16.607786178588867,2608.585125207901,0.18740869537812951,0.7708524827917046
|
| 62 |
+
60,0.08580842016424839,0.7053066492080688,30807,30807,79,37062,10,7.979757383000106,4.149253731343284,5e-06,0.12250088769652615,,0.12250088769652615,1.0,114,7.023168981000708,,,,,,,,,,,,,,,,,,,,,,,,,17.10348391532898,2625.68860912323,0.1892270434333476,0.7685060148830153
|
| 63 |
+
61,0.0790212529790082,0.708297073841095,30631,30631,78,37121,10,7.996359284001301,4.222014925373134,5e-06,0.12280233464186377,,0.12280233464186377,1.0,114,7.006872150001072,,,,,,,,,,,,,,,,,,,,,,,,,16.543776750564575,2642.2323858737946,0.19106860337647935,0.7663460535479708
|
| 64 |
+
62,0.0768654361886733,0.6585372686386108,31695,31695,73,37268,10,8.008203065000998,4.290111940298507,5e-06,0.1231296727358807,,0.1231296727358807,1.0,114,7.009335991000626,,,,,,,,,,,,,,,,,,,,,,,,,16.600826025009155,2658.8332118988037,0.19288756155703132,0.7641975015089274
|
| 65 |
+
63,0.081571021704259,0.7791351675987244,30547,30547,79,37151,10,7.992610985000283,4.3638059701492535,5e-06,0.12302364231747973,,0.12302364231747973,1.0,114,7.034464910000679,,,,,,,,,,,,,,,,,,,,,,,,,17.102600812911987,2675.9358127117157,0.19464161410030137,0.7619421036507795
|
| 66 |
+
64,0.08072052331389167,0.6853432655334473,31033,31033,82,37353,10,7.987528347999614,4.440298507462686,5e-06,0.12320919554968142,,0.12320919554968142,1.0,114,7.008643822999147,1.461883544921875,11.902786254882812,93.52164959907532,0.378125,0.35,0.8,0.0,523.640625,1043.6000000000001,0.046875,442.6885245901639,573.520202020202,90.92790007591248,0.23125,0.25,0.625,0.0,516.36875,1235.0000000000007,0.075,430.1216216216216,542.3130081300814,0.5905776023864746,198.9851393050012,215.5794780254364,2891.515290737152,0.18289233878759473,0.7763756092494039
|
| 67 |
+
65,0.08817977081539378,0.7155236601829529,30194,30194,87,37417,10,8.385498304000066,4.521455223880597,5e-06,0.12330166393782178,,0.12330166393782178,1.0,114,7.007902979999926,,,,,,,,,,,,,,,,,,,,,,,,,17.069241046905518,2908.5845317840576,0.18470203860208617,0.7742287782500015
|
| 68 |
+
66,0.0824841282760866,0.6839224696159363,30715,30715,79,37362,10,7.985244528999829,4.595149253731344,5e-06,0.12317220819442526,,0.12317220819442526,1.0,114,7.034345672998825,,,,,,,,,,,,,,,,,,,,,,,,,17.15778684616089,2925.7423186302185,0.1863481734274076,0.7720926685063025
|
| 69 |
+
67,0.06638792010145847,0.6320986747741699,31540,31540,65,36680,10,8.03044178700111,4.655783582089552,5e-06,0.12373688181800249,,0.12373688181800249,1.0,114,7.008368169999812,,,,,,,,,,,,,,,,,,,,,,,,,16.608852863311768,2942.3511714935303,0.18802554370224572,0.7701162880475718
|
| 70 |
+
68,0.07098985637621366,0.7079245448112488,32237,32237,66,37305,10,7.987338384000395,4.717350746268656,5e-06,0.12424052630540706,,0.12424052630540706,1.0,114,7.007636774000275,,,,,,,,,,,,,,,,,,,,,,,,,16.638245105743408,2958.9894165992737,0.18966763247569124,0.7681542172291054
|
| 71 |
+
69,0.07372805042773525,0.6848422884941101,31094,31094,80,37269,10,7.96994578200065,4.791977611940299,5e-06,0.12441683269879471,,0.12441683269879471,1.0,114,7.0362790190010855,,,,,,,,,,,,,,,,,,,,,,,,,17.07629370689392,2976.0657103061676,0.19125735731502208,0.7661109330416851
|
| 72 |
+
70,0.07699754901960784,0.6463869214057922,30600,30600,72,36949,10,8.012635647000934,4.8591417910447765,5e-06,0.1252613773104768,,0.1252613773104768,1.0,114,7.012893979999717,,,,,,,,,,,,,,,,,,,,,,,,,16.691897869110107,2992.7576081752777,0.19286797467969455,0.7641812908097297
|
| 73 |
+
71,0.07894779376111201,0.6815280914306641,30935,30935,76,37414,10,8.025957784999264,4.930037313432836,5e-06,0.12549193182490678,,0.12549193182490678,1.0,114,7.013776677000351,1.4668023586273193,12.465062379837036,89.33240604400635,0.415625,0.425,0.75,0.0,450.665625,888.6000000000001,0.025,430.18045112781954,465.2352941176471,88.1545934677124,0.215625,0.125,0.65,0.0,453.196875,870.6000000000008,0.05625,369.84057971014494,476.1115537848606,0.48834872245788574,192.49295615499977,209.1990008354187,3201.9566090106964,0.1827735749866487,0.7765614618045302
|
runs/fo2_sim4/run_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"args": {
|
| 3 |
+
"run_name": "fo2_sim4",
|
| 4 |
+
"pool": "arm2_sim4.parquet",
|
| 5 |
+
"val_pool": "../2026-07-06_0719pm_fixed-opener-training/ours_val.parquet",
|
| 6 |
+
"lr": 5e-06,
|
| 7 |
+
"seed": 0,
|
| 8 |
+
"seqs_per_step": 96,
|
| 9 |
+
"step_token_budget": null,
|
| 10 |
+
"buckets": "4096",
|
| 11 |
+
"max_num_docs": 32,
|
| 12 |
+
"max_comp_len": 2064,
|
| 13 |
+
"n_epochs": 10,
|
| 14 |
+
"max_steps": 72,
|
| 15 |
+
"start": 0,
|
| 16 |
+
"no_compile": false,
|
| 17 |
+
"resume": false,
|
| 18 |
+
"keep_every": 3,
|
| 19 |
+
"holdout_every": 1,
|
| 20 |
+
"holdout_ce_rollouts": 128,
|
| 21 |
+
"gen_holdout": 8,
|
| 22 |
+
"k_val": 8,
|
| 23 |
+
"gen_max_tokens": 2048,
|
| 24 |
+
"kv_pool_gb": 55.0,
|
| 25 |
+
"max_num_seqs": 512,
|
| 26 |
+
"no_graphs": false,
|
| 27 |
+
"no_wandb": false,
|
| 28 |
+
"no_push": false,
|
| 29 |
+
"project": "guided-rl-fixed-opener"
|
| 30 |
+
},
|
| 31 |
+
"model": "Qwen/Qwen3-8B-Base",
|
| 32 |
+
"git_sha": "74ca385",
|
| 33 |
+
"started_utc": "2026-07-07T20:06:37Z",
|
| 34 |
+
"inputs": {
|
| 35 |
+
"arm2_sim4.parquet": "e6b004e01d7f1694",
|
| 36 |
+
"../2026-07-06_0719pm_fixed-opener-training/ours_val.parquet": "9a83953d3559f597",
|
| 37 |
+
"/home/ubuntu/agent-ops/guided-rl/2026-07-06_0719pm_fixed-opener-training/split_train_val.json": "ba3543e8ca7fa086"
|
| 38 |
+
}
|
| 39 |
+
}
|
runs/fo2_sim4/state.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"step": 71,
|
| 3 |
+
"updated_utc": "2026-07-07T21:00:01Z",
|
| 4 |
+
"git_sha": "74ca385"
|
| 5 |
+
}
|
runs/fo2_sim4/wandb_run_id.txt
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
bwkaqix7
|