Auto upload zain 2026-08-14T20:23:04.132662 (part 2)

.gitattributes

@@ -45,3 +45,4 @@ zain/Activation/wandb/run-20260814_193907-kk1xbqht/run-kk1xbqht.wandb filter=lfs
 zain/Activation/wandb/run-20260814_195151-ix6wcbwv/run-ix6wcbwv.wandb filter=lfs diff=lfs merge=lfs -text
 zain/Activation/wandb/run-20260814_200425-npumf6lz/run-npumf6lz.wandb filter=lfs diff=lfs merge=lfs -text
 zain/Activation/wandb/run-20260814_200503-bdhnr27j/run-bdhnr27j.wandb filter=lfs diff=lfs merge=lfs -text
+zain/Activation/wandb/run-20260814_201615-w8mr4w6k/run-w8mr4w6k.wandb filter=lfs diff=lfs merge=lfs -text

zain/Activation/out/glu-waleed10-100L_run/checkpoint-100/trainer_state.json

@@ -3,7 +3,7 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 0.006740815638692282,
-  "eval_steps": 50,
+  "eval_steps": 2498,
   "global_step": 100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
@@ -13,51 +13,35 @@
       "epoch": 0.0013481631277384564,
       "grad_norm": 2.125,
       "learning_rate": 2.66e-05,
-      "loss": 8.29942626953125,
+      "loss": 8.29943389892578,
       "step": 20
     },
     {
       "epoch": 0.002696326255476913,
       "grad_norm": 1.3046875,
       "learning_rate": 5.46e-05,
-      "loss": 8.036360168457032,
+      "loss": 8.03638916015625,
       "step": 40
     },
-    {
-      "epoch": 0.003370407819346141,
-      "eval_loss": 7.832070827484131,
-      "eval_runtime": 15.4822,
-      "eval_samples_per_second": 615.352,
-      "eval_steps_per_second": 0.646,
-      "step": 50
-    },
     {
       "epoch": 0.004044489383215369,
       "grad_norm": 1.2734375,
       "learning_rate": 8.259999999999999e-05,
-      "loss": 7.834564208984375,
+      "loss": 7.834577941894532,
       "step": 60
     },
     {
       "epoch": 0.005392652510953826,
       "grad_norm": 1.265625,
       "learning_rate": 0.0001106,
-      "loss": 7.5718849182128904,
+      "loss": 7.5719154357910154,
       "step": 80
     },
     {
       "epoch": 0.006740815638692282,
       "grad_norm": 1.1953125,
       "learning_rate": 0.0001386,
-      "loss": 7.2372383117675785,
-      "step": 100
-    },
-    {
-      "epoch": 0.006740815638692282,
-      "eval_loss": 7.0443572998046875,
-      "eval_runtime": 15.5098,
-      "eval_samples_per_second": 614.257,
-      "eval_steps_per_second": 0.645,
+      "loss": 7.23726806640625,
       "step": 100
     }
   ],

zain/Activation/out/glu-waleed10-100L_run/checkpoint-100/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95042bce3268a58e39bf6898ae4dc3d7f41ae0258221813fdf6720b6f7599e44
+oid sha256:9a5eecd63c938d7d339a59fa838b1d3989a5cd93b45f48430293465bdb710244
 size 4920

zain/Activation/out/glu-waleed10-100L_run/checkpoint-200/config.json

@@ -32,5 +32,5 @@
   "transformers_version": "5.16.0.dev0",
   "use_cache": false,
   "vocab_size": 4096,
-  "waleed_beta": 10.0
+  "waleed_beta": 4
 }

zain/Activation/out/glu-waleed10-100L_run/checkpoint-200/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d1fa8cb579eecffad489fd4cec7e2a3d9d3fe35881a7cc2399dc4ac53186207
+oid sha256:1d298f14a52352373baec0551092bb19d6b60ea754318000f80bd7af3acd6610
 size 33967272

zain/Activation/out/glu-waleed10-100L_run/checkpoint-200/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c59b0e8734f31cdf307508221b8cd33e3928c5a4fc4e7ec93316f115d367fdc4
+oid sha256:ad879b16b566df10c0ea2a2cd7f1faccc93b3ee8ed7b31e5788bde05acf7b6dd
 size 68504996

zain/Activation/out/glu-waleed10-100L_run/checkpoint-200/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ecefbb3f17bb76b6655eb0157c98b5287c17fa4b4c72a6b9068b0823ce9fd18d
+oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
 size 14244

zain/Activation/out/glu-waleed10-100L_run/checkpoint-200/trainer_state.json

@@ -3,7 +3,7 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 0.013481631277384564,
-  "eval_steps": 50,
+  "eval_steps": 2498,
   "global_step": 200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
@@ -13,102 +13,70 @@
       "epoch": 0.0013481631277384564,
       "grad_norm": 2.125,
       "learning_rate": 2.66e-05,
-      "loss": 8.29942626953125,
+      "loss": 8.29943389892578,
       "step": 20
     },
     {
       "epoch": 0.002696326255476913,
       "grad_norm": 1.3046875,
       "learning_rate": 5.46e-05,
-      "loss": 8.036360168457032,
+      "loss": 8.03638916015625,
       "step": 40
     },
-    {
-      "epoch": 0.003370407819346141,
-      "eval_loss": 7.832070827484131,
-      "eval_runtime": 15.4822,
-      "eval_samples_per_second": 615.352,
-      "eval_steps_per_second": 0.646,
-      "step": 50
-    },
     {
       "epoch": 0.004044489383215369,
       "grad_norm": 1.2734375,
       "learning_rate": 8.259999999999999e-05,
-      "loss": 7.834564208984375,
+      "loss": 7.834577941894532,
       "step": 60
     },
     {
       "epoch": 0.005392652510953826,
       "grad_norm": 1.265625,
       "learning_rate": 0.0001106,
-      "loss": 7.5718849182128904,
+      "loss": 7.5719154357910154,
       "step": 80
     },
     {
       "epoch": 0.006740815638692282,
       "grad_norm": 1.1953125,
       "learning_rate": 0.0001386,
-      "loss": 7.2372383117675785,
-      "step": 100
-    },
-    {
-      "epoch": 0.006740815638692282,
-      "eval_loss": 7.0443572998046875,
-      "eval_runtime": 15.5098,
-      "eval_samples_per_second": 614.257,
-      "eval_steps_per_second": 0.645,
+      "loss": 7.23726806640625,
       "step": 100
     },
     {
       "epoch": 0.008088978766430738,
       "grad_norm": 1.1484375,
       "learning_rate": 0.00016659999999999998,
-      "loss": 6.864932250976563,
+      "loss": 6.864923095703125,
       "step": 120
     },
     {
       "epoch": 0.009437141894169195,
       "grad_norm": 1.0234375,
       "learning_rate": 0.00019460000000000001,
-      "loss": 6.509996032714843,
+      "loss": 6.509650421142578,
       "step": 140
     },
-    {
-      "epoch": 0.010111223458038422,
-      "eval_loss": 6.1924004554748535,
-      "eval_runtime": 15.382,
-      "eval_samples_per_second": 619.361,
-      "eval_steps_per_second": 0.65,
-      "step": 150
-    },
     {
       "epoch": 0.010785305021907651,
-      "grad_norm": 0.96484375,
+      "grad_norm": 1.03125,
       "learning_rate": 0.0002226,
-      "loss": 6.203652191162109,
+      "loss": 6.204092788696289,
       "step": 160
     },
     {
       "epoch": 0.012133468149646108,
-      "grad_norm": 1.0234375,
+      "grad_norm": 0.87109375,
       "learning_rate": 0.00025059999999999997,
-      "loss": 5.938727951049804,
+      "loss": 5.9408447265625,
       "step": 180
     },
     {
       "epoch": 0.013481631277384564,
-      "grad_norm": 0.90234375,
+      "grad_norm": 0.76171875,
       "learning_rate": 0.0002786,
-      "loss": 5.677180480957031,
-      "step": 200
-    },
-    {
-      "epoch": 0.013481631277384564,
-      "eval_loss": 5.540520668029785,
-      "eval_runtime": 15.6734,
-      "eval_samples_per_second": 607.844,
-      "eval_steps_per_second": 0.638,
+      "loss": 5.677399826049805,
       "step": 200
     }
   ],

zain/Activation/out/glu-waleed10-100L_run/checkpoint-200/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95042bce3268a58e39bf6898ae4dc3d7f41ae0258221813fdf6720b6f7599e44
+oid sha256:9a5eecd63c938d7d339a59fa838b1d3989a5cd93b45f48430293465bdb710244
 size 4920

zain/Activation/out/glu-waleed10-100L_run/training_log.jsonl

@@ -1,177 +1,13 @@
-{"step": 20, "epoch": 0.0013481631277384564, "timestamp": 1786572099.4953449, "loss": 8.29942626953125, "grad_norm": 2.125, "learning_rate": 2.66e-05, "train/total_time_seconds": 5.507525555789471, "train/time_per_step_avg": 0.27537627778947354, "train/epoch_time_elapsed": 6.568354923278093, "train/estimated_remaining_minutes": 11.382219481964906}
-{"step": 40, "epoch": 0.002696326255476913, "timestamp": 1786572105.373673, "loss": 8.036360168457032, "grad_norm": 1.3046875, "learning_rate": 5.46e-05, "train/total_time_seconds": 10.417302578687668, "train/time_per_step_avg": 0.2604325644671917, "train/epoch_time_elapsed": 12.446682956069708, "train/estimated_remaining_minutes": 10.677735143154859}
-{"step": 50, "epoch": 0.003370407819346141, "timestamp": 1786572123.7892528, "eval_loss": 7.832070827484131, "eval_runtime": 15.4822, "eval_samples_per_second": 615.352, "eval_steps_per_second": 0.646, "train/total_time_seconds": 12.863984804600477, "train/time_per_step_avg": 0.25727969609200957, "train/epoch_time_elapsed": 30.862261060625315, "train/estimated_remaining_minutes": 10.505587590423724}
-{"step": 60, "epoch": 0.004044489383215369, "timestamp": 1786572126.7337923, "loss": 7.834564208984375, "grad_norm": 1.2734375, "learning_rate": 8.259999999999999e-05, "train/total_time_seconds": 15.315555311739445, "train/time_per_step_avg": 0.2552592551956574, "train/epoch_time_elapsed": 33.80680202320218, "train/estimated_remaining_minutes": 10.380543044623401}
-{"step": 80, "epoch": 0.005392652510953826, "timestamp": 1786572132.6449795, "loss": 7.5718849182128904, "grad_norm": 1.265625, "learning_rate": 0.0001106, "train/total_time_seconds": 20.22891864180565, "train/time_per_step_avg": 0.2528614830225706, "train/epoch_time_elapsed": 39.71798903495073, "train/estimated_remaining_minutes": 10.198746481910348}
-{"step": 100, "epoch": 0.006740815638692282, "timestamp": 1786572138.4929912, "loss": 7.2372383117675785, "grad_norm": 1.1953125, "learning_rate": 0.0001386, "train/total_time_seconds": 25.11151283606887, "train/time_per_step_avg": 0.25111512836068867, "train/epoch_time_elapsed": 45.56600131466985, "train/estimated_remaining_minutes": 10.044605134427547}
-{"step": 100, "epoch": 0.006740815638692282, "timestamp": 1786572154.0043495, "eval_loss": 7.0443572998046875, "eval_runtime": 15.5098, "eval_samples_per_second": 614.257, "eval_steps_per_second": 0.645, "train/total_time_seconds": 25.11151283606887, "train/time_per_step_avg": 0.25111512836068867, "train/epoch_time_elapsed": 61.07735696434975, "train/estimated_remaining_minutes": 10.044605134427547}
-{"step": 120, "epoch": 0.008088978766430738, "timestamp": 1786572160.1766303, "loss": 6.864932250976563, "grad_norm": 1.1484375, "learning_rate": 0.00016659999999999998, "train/total_time_seconds": 30.000421818345785, "train/time_per_step_avg": 0.24492896262556316, "train/epoch_time_elapsed": 67.24964017421007, "train/estimated_remaining_minutes": 9.916806101064301}
-{"step": 140, "epoch": 0.009437141894169195, "timestamp": 1786572166.0242524, "loss": 6.509996032714843, "grad_norm": 1.0234375, "learning_rate": 0.00019460000000000001, "train/total_time_seconds": 34.87341859191656, "train/time_per_step_avg": 0.24456116013228893, "train/epoch_time_elapsed": 73.09726232662797, "train/estimated_remaining_minutes": 9.797769985347987}
-{"step": 150, "epoch": 0.010111223458038422, "timestamp": 1786572184.3254743, "eval_loss": 6.1924004554748535, "eval_runtime": 15.382, "eval_samples_per_second": 619.361, "eval_steps_per_second": 0.65, "train/total_time_seconds": 37.30649349093437, "train/time_per_step_avg": 0.24442508686333894, "train/epoch_time_elapsed": 91.39848183095455, "train/estimated_remaining_minutes": 9.74113996707731}
-{"step": 160, "epoch": 0.010785305021907651, "timestamp": 1786572187.291386, "loss": 6.203652191162109, "grad_norm": 0.96484375, "learning_rate": 0.0002226, "train/total_time_seconds": 39.77944863587618, "train/time_per_step_avg": 0.24463893324136735, "train/epoch_time_elapsed": 94.36439569294453, "train/estimated_remaining_minutes": 9.696240604994818}
-{"step": 180, "epoch": 0.012133468149646108, "timestamp": 1786572193.1732507, "loss": 5.938727951049804, "grad_norm": 1.0234375, "learning_rate": 0.00025059999999999997, "train/total_time_seconds": 44.6577689871192, "train/time_per_step_avg": 0.2442885034531355, "train/epoch_time_elapsed": 100.24625926464796, "train/estimated_remaining_minutes": 9.59315037501079}
-{"step": 200, "epoch": 0.013481631277384564, "timestamp": 1786572199.0488183, "loss": 5.677180480957031, "grad_norm": 0.90234375, "learning_rate": 0.0002786, "train/total_time_seconds": 49.54687497764826, "train/time_per_step_avg": 0.2443536214157939, "train/epoch_time_elapsed": 106.12182818353176, "train/estimated_remaining_minutes": 9.496484370715915}
-{"step": 200, "epoch": 0.013481631277384564, "timestamp": 1786572214.7237773, "eval_loss": 5.540520668029785, "eval_runtime": 15.6734, "eval_samples_per_second": 607.844, "eval_steps_per_second": 0.638, "train/total_time_seconds": 49.54687497764826, "train/time_per_step_avg": 0.2443536214157939, "train/epoch_time_elapsed": 121.79678123444319, "train/estimated_remaining_minutes": 9.496484370715915}
-{"step": 220, "epoch": 0.01482979440512302, "timestamp": 1786572220.86287, "loss": 5.418574523925781, "grad_norm": 1.5078125, "learning_rate": 0.00030659999999999997, "train/total_time_seconds": 54.390119556337595, "train/time_per_step_avg": 0.2438969773799181, "train/epoch_time_elapsed": 127.93587946891785, "train/estimated_remaining_minutes": 9.394657014276495}
-{"step": 240, "epoch": 0.016177957532861477, "timestamp": 1786572226.7335033, "loss": 5.1351776123046875, "grad_norm": 0.7109375, "learning_rate": 0.0003346, "train/total_time_seconds": 59.29074428230524, "train/time_per_step_avg": 0.2441732569038868, "train/epoch_time_elapsed": 133.80651319026947, "train/estimated_remaining_minutes": 9.305352922084017}
-{"step": 250, "epoch": 0.016852039096730706, "timestamp": 1786572245.075726, "eval_loss": 4.857514381408691, "eval_runtime": 15.4191, "eval_samples_per_second": 617.872, "eval_steps_per_second": 0.649, "train/total_time_seconds": 61.729648772627115, "train/time_per_step_avg": 0.24423155281692743, "train/epoch_time_elapsed": 152.1487331353128, "train/estimated_remaining_minutes": 9.259447315894068}
-{"step": 260, "epoch": 0.01752612066059993, "timestamp": 1786572247.9988925, "loss": 4.880479431152343, "grad_norm": 0.875, "learning_rate": 0.00036260000000000003, "train/total_time_seconds": 64.1646473146975, "train/time_per_step_avg": 0.24385198678821327, "train/epoch_time_elapsed": 155.07190223783255, "train/estimated_remaining_minutes": 9.213385255443745}
-{"step": 280, "epoch": 0.01887428378833839, "timestamp": 1786572253.9181268, "loss": 4.645774841308594, "grad_norm": 0.875, "learning_rate": 0.0003906, "train/total_time_seconds": 69.10982979089022, "train/time_per_step_avg": 0.2445206080377102, "train/epoch_time_elapsed": 160.99113485962152, "train/estimated_remaining_minutes": 9.132370365224778}
-{"step": 300, "epoch": 0.020222446916076844, "timestamp": 1786572259.8048985, "loss": 4.451469421386719, "grad_norm": 0.6640625, "learning_rate": 0.0004186, "train/total_time_seconds": 73.9895646609366, "train/time_per_step_avg": 0.24442689683288335, "train/epoch_time_elapsed": 166.87790881842375, "train/estimated_remaining_minutes": 9.043169014114474}
-{"step": 300, "epoch": 0.020222446916076844, "timestamp": 1786572275.200316, "eval_loss": 4.3544721603393555, "eval_runtime": 15.3937, "eval_samples_per_second": 618.891, "eval_steps_per_second": 0.65, "train/total_time_seconds": 73.9895646609366, "train/time_per_step_avg": 0.24442689683288335, "train/epoch_time_elapsed": 182.27332329005003, "train/estimated_remaining_minutes": 9.043169014114474}
-{"step": 320, "epoch": 0.021570610043815303, "timestamp": 1786572281.3944638, "loss": 4.269550323486328, "grad_norm": 0.609375, "learning_rate": 0.0004466, "train/total_time_seconds": 78.85832178965211, "train/time_per_step_avg": 0.24468202233314515, "train/epoch_time_elapsed": 188.46747362613678, "train/estimated_remaining_minutes": 8.953705286533417}
-{"step": 340, "epoch": 0.022918773171553757, "timestamp": 1786572287.2559514, "loss": 4.115790176391601, "grad_norm": 0.50390625, "learning_rate": 0.00047460000000000004, "train/total_time_seconds": 83.7492892332375, "train/time_per_step_avg": 0.24458544950932265, "train/epoch_time_elapsed": 194.32896095886827, "train/estimated_remaining_minutes": 8.867571801166324}
-{"step": 350, "epoch": 0.023592854735422986, "timestamp": 1786572305.5230896, "eval_loss": 3.9874157905578613, "eval_runtime": 15.3586, "eval_samples_per_second": 620.304, "eval_steps_per_second": 0.651, "train/total_time_seconds": 86.17307840287685, "train/time_per_step_avg": 0.24443429630249738, "train/epoch_time_elapsed": 212.59609816223383, "train/estimated_remaining_minutes": 8.822481836485013}
-{"step": 360, "epoch": 0.024266936299292215, "timestamp": 1786572308.4556365, "loss": 3.9873260498046874, "grad_norm": 0.8828125, "learning_rate": 0.0005026, "train/total_time_seconds": 88.60832645744085, "train/time_per_step_avg": 0.2444367914274335, "train/epoch_time_elapsed": 215.5286463163793, "train/estimated_remaining_minutes": 8.778787899024232}
-{"step": 380, "epoch": 0.02561509942703067, "timestamp": 1786572314.518654, "loss": 3.873243713378906, "grad_norm": 0.65234375, "learning_rate": 0.0005306, "train/total_time_seconds": 93.69690709933639, "train/time_per_step_avg": 0.2458707730844617, "train/epoch_time_elapsed": 221.59166384115815, "train/estimated_remaining_minutes": 8.712168554850575}
-{"step": 400, "epoch": 0.026963262554769128, "timestamp": 1786572320.4469628, "loss": 3.7449169158935547, "grad_norm": 0.50390625, "learning_rate": 0.0005586, "train/total_time_seconds": 98.6364342905581, "train/time_per_step_avg": 0.24646869629621507, "train/epoch_time_elapsed": 227.51997267827392, "train/estimated_remaining_minutes": 8.630688000423834}
-{"step": 400, "epoch": 0.026963262554769128, "timestamp": 1786572335.8352714, "eval_loss": 3.713116407394409, "eval_runtime": 15.3868, "eval_samples_per_second": 619.166, "eval_steps_per_second": 0.65, "train/total_time_seconds": 98.6364342905581, "train/time_per_step_avg": 0.24646869629621507, "train/epoch_time_elapsed": 242.90827917307615, "train/estimated_remaining_minutes": 8.630688000423834}
-{"step": 420, "epoch": 0.028311425682507583, "timestamp": 1786572341.998572, "loss": 3.653180694580078, "grad_norm": 0.53125, "learning_rate": 0.0005866, "train/total_time_seconds": 103.52565556392074, "train/time_per_step_avg": 0.24667333774268627, "train/epoch_time_elapsed": 249.07158163934946, "train/estimated_remaining_minutes": 8.544974744958536}
-{"step": 440, "epoch": 0.02965958881024604, "timestamp": 1786572347.904592, "loss": 3.599313735961914, "grad_norm": 0.59375, "learning_rate": 0.0006146, "train/total_time_seconds": 108.3871761970222, "train/time_per_step_avg": 0.24637886963784694, "train/epoch_time_elapsed": 254.97760132327676, "train/estimated_remaining_minutes": 8.45748420325249}
-{"step": 450, "epoch": 0.030333670374115267, "timestamp": 1786572366.1972334, "eval_loss": 3.5332272052764893, "eval_runtime": 15.3821, "eval_samples_per_second": 619.355, "eval_steps_per_second": 0.65, "train/total_time_seconds": 110.8073519654572, "train/time_per_step_avg": 0.24634273562580347, "train/epoch_time_elapsed": 273.2702407948673, "train/estimated_remaining_minutes": 8.413150797377305}
-{"step": 460, "epoch": 0.031007751937984496, "timestamp": 1786572369.1516082, "loss": 3.5110404968261717, "grad_norm": 0.57421875, "learning_rate": 0.0006426, "train/total_time_seconds": 113.25167169421911, "train/time_per_step_avg": 0.24643345236778258, "train/epoch_time_elapsed": 276.22461833804846, "train/estimated_remaining_minutes": 8.370775733920542}
-{"step": 480, "epoch": 0.032355915065722954, "timestamp": 1786572374.984092, "loss": 3.4395523071289062, "grad_norm": 0.671875, "learning_rate": 0.0006705999999999999, "train/total_time_seconds": 118.11047495529056, "train/time_per_step_avg": 0.2441356785595417, "train/epoch_time_elapsed": 282.0571016408503, "train/estimated_remaining_minutes": 8.284137479503018}
-{"step": 500, "epoch": 0.03370407819346141, "timestamp": 1786572380.8576894, "loss": 3.3704925537109376, "grad_norm": 0.53125, "learning_rate": 0.0006986, "train/total_time_seconds": 123.00776559114456, "train/time_per_step_avg": 0.24371331300586463, "train/epoch_time_elapsed": 287.93069849535823, "train/estimated_remaining_minutes": 8.200517706076305}
-{"step": 500, "epoch": 0.03370407819346141, "timestamp": 1786572396.3801348, "eval_loss": 3.340529680252075, "eval_runtime": 15.5207, "eval_samples_per_second": 613.824, "eval_steps_per_second": 0.644, "train/total_time_seconds": 123.00776559114456, "train/time_per_step_avg": 0.24371331300586463, "train/epoch_time_elapsed": 303.45314341038465, "train/estimated_remaining_minutes": 8.200517706076305}
-{"step": 520, "epoch": 0.03505224132119986, "timestamp": 1786572402.5530741, "loss": 3.3107833862304688, "grad_norm": 0.49609375, "learning_rate": 0.0007, "train/total_time_seconds": 127.89689143747091, "train/time_per_step_avg": 0.24371235873550176, "train/epoch_time_elapsed": 309.6260833404958, "train/estimated_remaining_minutes": 8.11653349507027}
-{"step": 540, "epoch": 0.03640040444893832, "timestamp": 1786572408.4176164, "loss": 3.2391616821289064, "grad_norm": 0.51171875, "learning_rate": 0.0007, "train/total_time_seconds": 132.78322450071573, "train/time_per_step_avg": 0.24396048303693532, "train/epoch_time_elapsed": 315.4906254261732, "train/estimated_remaining_minutes": 8.032565432759347}
-{"step": 550, "epoch": 0.03707448601280755, "timestamp": 1786572427.0445926, "eval_loss": 3.2028048038482666, "eval_runtime": 15.6434, "eval_samples_per_second": 609.013, "eval_steps_per_second": 0.639, "train/total_time_seconds": 135.232557002455, "train/time_per_step_avg": 0.24425205036997796, "train/epoch_time_elapsed": 334.1176012568176, "train/estimated_remaining_minutes": 7.99101473196325}
-{"step": 560, "epoch": 0.03774856757667678, "timestamp": 1786572429.9867537, "loss": 3.1940614700317385, "grad_norm": 0.466796875, "learning_rate": 0.0007, "train/total_time_seconds": 137.6735647507012, "train/time_per_step_avg": 0.24421893056482077, "train/epoch_time_elapsed": 337.05976240336895, "train/estimated_remaining_minutes": 7.949009393344057}
-{"step": 580, "epoch": 0.03909673070441524, "timestamp": 1786572435.8094902, "loss": 3.141546440124512, "grad_norm": 0.4921875, "learning_rate": 0.0007, "train/total_time_seconds": 142.52040734142065, "train/time_per_step_avg": 0.24409932386130095, "train/epoch_time_elapsed": 342.88249949365854, "train/estimated_remaining_minutes": 7.863194887802519}
-{"step": 600, "epoch": 0.04044489383215369, "timestamp": 1786572441.640344, "loss": 3.1114213943481444, "grad_norm": 0.4609375, "learning_rate": 0.0007, "train/total_time_seconds": 147.37215169146657, "train/time_per_step_avg": 0.24364386100322008, "train/epoch_time_elapsed": 348.7133532837033, "train/estimated_remaining_minutes": 7.777974672605181}
-{"step": 600, "epoch": 0.04044489383215369, "timestamp": 1786572457.046109, "eval_loss": 3.0854389667510986, "eval_runtime": 15.4041, "eval_samples_per_second": 618.47, "eval_steps_per_second": 0.649, "train/total_time_seconds": 147.37215169146657, "train/time_per_step_avg": 0.24364386100322008, "train/epoch_time_elapsed": 364.11911669000983, "train/estimated_remaining_minutes": 7.777974672605181}
-{"step": 620, "epoch": 0.04179305695989215, "timestamp": 1786572463.1804712, "loss": 3.0545623779296873, "grad_norm": 0.5546875, "learning_rate": 0.0007, "train/total_time_seconds": 152.22994845360518, "train/time_per_step_avg": 0.24333057016134263, "train/epoch_time_elapsed": 370.2534808591008, "train/estimated_remaining_minutes": 7.693341480988648}
-{"step": 640, "epoch": 0.043141220087630605, "timestamp": 1786572469.001045, "loss": 3.0375244140625, "grad_norm": 0.50390625, "learning_rate": 0.0007, "train/total_time_seconds": 157.06764130666852, "train/time_per_step_avg": 0.24284416805952788, "train/epoch_time_elapsed": 376.07405441999435, "train/estimated_remaining_minutes": 7.607963875791756}
-{"step": 650, "epoch": 0.043815301651499834, "timestamp": 1786572487.3125155, "eval_loss": 2.997983455657959, "eval_runtime": 15.3695, "eval_samples_per_second": 619.864, "eval_steps_per_second": 0.651, "train/total_time_seconds": 159.51168559119105, "train/time_per_step_avg": 0.24279128588736057, "train/epoch_time_elapsed": 394.3855227380991, "train/estimated_remaining_minutes": 7.566579957530858}
-{"step": 660, "epoch": 0.044489383215369056, "timestamp": 1786572490.2232594, "loss": 2.9873325347900392, "grad_norm": 0.470703125, "learning_rate": 0.0007, "train/total_time_seconds": 161.93452242016792, "train/time_per_step_avg": 0.24260957669466734, "train/epoch_time_elapsed": 397.29626908525825, "train/estimated_remaining_minutes": 7.52423033467447}
-{"step": 680, "epoch": 0.045837546343107514, "timestamp": 1786572496.0884476, "loss": 2.9647268295288085, "grad_norm": 0.462890625, "learning_rate": 0.0007, "train/total_time_seconds": 166.81659445539117, "train/time_per_step_avg": 0.2429618711397052, "train/epoch_time_elapsed": 403.1614573597908, "train/estimated_remaining_minutes": 7.441328478157155}
-{"step": 700, "epoch": 0.04718570947084597, "timestamp": 1786572501.9507153, "loss": 2.924059295654297, "grad_norm": 0.435546875, "learning_rate": 0.0007, "train/total_time_seconds": 171.6934934668243, "train/time_per_step_avg": 0.24321341775357724, "train/epoch_time_elapsed": 409.02372374385595, "train/estimated_remaining_minutes": 7.358292577149612}
-{"step": 700, "epoch": 0.04718570947084597, "timestamp": 1786572517.3963296, "eval_loss": 2.9152462482452393, "eval_runtime": 15.4441, "eval_samples_per_second": 616.871, "eval_steps_per_second": 0.647, "train/total_time_seconds": 171.6934934668243, "train/time_per_step_avg": 0.24321341775357724, "train/epoch_time_elapsed": 424.4693373963237, "train/estimated_remaining_minutes": 7.358292577149612}
-{"step": 720, "epoch": 0.04853387259858443, "timestamp": 1786572523.656865, "loss": 2.895351600646973, "grad_norm": 0.447265625, "learning_rate": 0.0007, "train/total_time_seconds": 176.629100240767, "train/time_per_step_avg": 0.24399151787161827, "train/epoch_time_elapsed": 430.72987451404333, "train/estimated_remaining_minutes": 7.277773111772344}
-{"step": 740, "epoch": 0.04988203572632288, "timestamp": 1786572529.7448912, "loss": 2.861538314819336, "grad_norm": 0.4453125, "learning_rate": 0.0007, "train/total_time_seconds": 181.7423182502389, "train/time_per_step_avg": 0.24674676943570376, "train/epoch_time_elapsed": 436.8179010003805, "train/estimated_remaining_minutes": 7.204200002712173}
-{"step": 750, "epoch": 0.05055611729019211, "timestamp": 1786572548.1168342, "eval_loss": 2.849472761154175, "eval_runtime": 15.4441, "eval_samples_per_second": 616.868, "eval_steps_per_second": 0.647, "train/total_time_seconds": 184.17681366950274, "train/time_per_step_avg": 0.24665128078311682, "train/epoch_time_elapsed": 455.1898415274918, "train/estimated_remaining_minutes": 7.1624316427028845}
-{"step": 760, "epoch": 0.05123019885406134, "timestamp": 1786572551.0368974, "loss": 2.848223114013672, "grad_norm": 0.478515625, "learning_rate": 0.0007, "train/total_time_seconds": 186.60259008407593, "train/time_per_step_avg": 0.24668067663908005, "train/epoch_time_elapsed": 458.10990741848946, "train/estimated_remaining_minutes": 7.120361990050267}
-{"step": 780, "epoch": 0.0525783619817998, "timestamp": 1786572556.9486566, "loss": 2.8442684173583985, "grad_norm": 0.498046875, "learning_rate": 0.0007, "train/total_time_seconds": 191.51008857041597, "train/time_per_step_avg": 0.24693494115024806, "train/epoch_time_elapsed": 464.0216665491462, "train/estimated_remaining_minutes": 7.03840496455375}
-{"step": 800, "epoch": 0.053926525109538256, "timestamp": 1786572562.822533, "loss": 2.802508735656738, "grad_norm": 0.421875, "learning_rate": 0.0007, "train/total_time_seconds": 196.40453672409058, "train/time_per_step_avg": 0.24711043257266282, "train/epoch_time_elapsed": 469.8955425620079, "train/estimated_remaining_minutes": 6.955994008978208}
-{"step": 800, "epoch": 0.053926525109538256, "timestamp": 1786572578.203275, "eval_loss": 2.7919585704803467, "eval_runtime": 15.379, "eval_samples_per_second": 619.481, "eval_steps_per_second": 0.65, "train/total_time_seconds": 196.40453672409058, "train/time_per_step_avg": 0.24711043257266282, "train/epoch_time_elapsed": 485.2762820236385, "train/estimated_remaining_minutes": 6.955994008978208}
-{"step": 820, "epoch": 0.05527468823727671, "timestamp": 1786572584.3627603, "loss": 2.791913604736328, "grad_norm": 0.46484375, "learning_rate": 0.0007, "train/total_time_seconds": 201.27779496088624, "train/time_per_step_avg": 0.24648694720119238, "train/epoch_time_elapsed": 491.4357697889209, "train/estimated_remaining_minutes": 6.872900315737579}
-{"step": 840, "epoch": 0.056622851365015166, "timestamp": 1786572590.28871, "loss": 2.764504241943359, "grad_norm": 0.455078125, "learning_rate": 0.0007, "train/total_time_seconds": 206.20850687474012, "train/time_per_step_avg": 0.24466188624501228, "train/epoch_time_elapsed": 497.3617196381092, "train/estimated_remaining_minutes": 6.79178812325533}
-{"step": 850, "epoch": 0.057296932928884395, "timestamp": 1786572608.7183652, "eval_loss": 2.733065366744995, "eval_runtime": 15.5198, "eval_samples_per_second": 613.863, "eval_steps_per_second": 0.644, "train/total_time_seconds": 208.63105216249824, "train/time_per_step_avg": 0.244542384929955, "train/epoch_time_elapsed": 515.7913730181754, "train/estimated_remaining_minutes": 6.749828158198472}
-{"step": 860, "epoch": 0.057971014492753624, "timestamp": 1786572611.642772, "loss": 2.723073387145996, "grad_norm": 0.50390625, "learning_rate": 0.0007, "train/total_time_seconds": 211.06512178108096, "train/time_per_step_avg": 0.24462531697005033, "train/epoch_time_elapsed": 518.7157817147672, "train/estimated_remaining_minutes": 6.708271312421953}
-{"step": 880, "epoch": 0.05931917762049208, "timestamp": 1786572617.6065154, "loss": 2.700708198547363, "grad_norm": 0.419921875, "learning_rate": 0.0007, "train/total_time_seconds": 215.98648421093822, "train/time_per_step_avg": 0.24476395640522242, "train/epoch_time_elapsed": 524.679524961859, "train/estimated_remaining_minutes": 6.62685803829015}
-{"step": 900, "epoch": 0.06066734074823053, "timestamp": 1786572623.4193249, "loss": 2.6873470306396485, "grad_norm": 0.46484375, "learning_rate": 0.0007, "train/total_time_seconds": 220.82203654944897, "train/time_per_step_avg": 0.2441749982535839, "train/epoch_time_elapsed": 530.4923346191645, "train/estimated_remaining_minutes": 6.542875157020711}
-{"step": 900, "epoch": 0.06066734074823053, "timestamp": 1786572638.8386846, "eval_loss": 2.682546615600586, "eval_runtime": 15.4179, "eval_samples_per_second": 617.919, "eval_steps_per_second": 0.649, "train/total_time_seconds": 220.82203654944897, "train/time_per_step_avg": 0.2441749982535839, "train/epoch_time_elapsed": 545.9116909950972, "train/estimated_remaining_minutes": 6.542875157020711}
-{"step": 920, "epoch": 0.06201550387596899, "timestamp": 1786572645.2134163, "loss": 2.663492965698242, "grad_norm": 0.478515625, "learning_rate": 0.0007, "train/total_time_seconds": 225.92413433641195, "train/time_per_step_avg": 0.24646339375525714, "train/epoch_time_elapsed": 552.2864259406924, "train/estimated_remaining_minutes": 6.466669062527733}
-{"step": 940, "epoch": 0.06336366700370745, "timestamp": 1786572651.2076764, "loss": 2.653563690185547, "grad_norm": 0.46875, "learning_rate": 0.0007, "train/total_time_seconds": 230.86760304495692, "train/time_per_step_avg": 0.246590961702168, "train/epoch_time_elapsed": 558.2806855365634, "train/estimated_remaining_minutes": 6.385699658690298}
-{"step": 950, "epoch": 0.06403774856757667, "timestamp": 1786572669.7537134, "eval_loss": 2.636122941970825, "eval_runtime": 15.6175, "eval_samples_per_second": 610.019, "eval_steps_per_second": 0.64, "train/total_time_seconds": 233.3007367104292, "train/time_per_step_avg": 0.24669684547930956, "train/epoch_time_elapsed": 576.8267201557755, "train/estimated_remaining_minutes": 6.34414284037132}
-{"step": 960, "epoch": 0.06471183013144591, "timestamp": 1786572672.6229477, "loss": 2.628633499145508, "grad_norm": 0.455078125, "learning_rate": 0.0007, "train/total_time_seconds": 235.6729480214417, "train/time_per_step_avg": 0.24607826240360736, "train/epoch_time_elapsed": 579.6959564350545, "train/estimated_remaining_minutes": 6.300978124184379}
-{"step": 980, "epoch": 0.06605999325918437, "timestamp": 1786572678.4017718, "loss": 2.6126628875732423, "grad_norm": 0.44140625, "learning_rate": 0.0007, "train/total_time_seconds": 240.45006656646729, "train/time_per_step_avg": 0.2446358235552907, "train/epoch_time_elapsed": 585.4747802354395, "train/estimated_remaining_minutes": 6.215716006480107}
-{"step": 1000, "epoch": 0.06740815638692282, "timestamp": 1786572684.1698456, "loss": 2.6122547149658204, "grad_norm": 0.4453125, "learning_rate": 0.0007, "train/total_time_seconds": 245.20241347700357, "train/time_per_step_avg": 0.24380376927554606, "train/epoch_time_elapsed": 591.2428545951843, "train/estimated_remaining_minutes": 6.130060336925089}
-{"step": 1000, "epoch": 0.06740815638692282, "timestamp": 1786572699.6720955, "eval_loss": 2.5941500663757324, "eval_runtime": 15.5006, "eval_samples_per_second": 614.621, "eval_steps_per_second": 0.645, "train/total_time_seconds": 245.20241347700357, "train/time_per_step_avg": 0.24380376927554606, "train/epoch_time_elapsed": 606.7451028190553, "train/estimated_remaining_minutes": 6.130060336925089}
-{"step": 1020, "epoch": 0.06875631951466127, "timestamp": 1786572705.6853964, "loss": 2.5827909469604493, "grad_norm": 0.416015625, "learning_rate": 0.0007, "train/total_time_seconds": 249.95135128125548, "train/time_per_step_avg": 0.2402721694484353, "train/epoch_time_elapsed": 612.7584060057998, "train/estimated_remaining_minutes": 6.044575161703564}
-{"step": 1040, "epoch": 0.07010448264239973, "timestamp": 1786572711.4487538, "loss": 2.569642448425293, "grad_norm": 0.478515625, "learning_rate": 0.0007, "train/total_time_seconds": 254.71493316441774, "train/time_per_step_avg": 0.23847330119460822, "train/epoch_time_elapsed": 618.5217633396387, "train/estimated_remaining_minutes": 5.9596763208341335}
-{"step": 1050, "epoch": 0.07077856420626896, "timestamp": 1786572729.821028, "eval_loss": 2.5616002082824707, "eval_runtime": 15.4489, "eval_samples_per_second": 616.678, "eval_steps_per_second": 0.647, "train/total_time_seconds": 257.1310795173049, "train/time_per_step_avg": 0.23830342806875707, "train/epoch_time_elapsed": 636.8940357416868, "train/estimated_remaining_minutes": 5.918096274604637}
-{"step": 1060, "epoch": 0.07145264577013818, "timestamp": 1786572732.6916323, "loss": 2.558013153076172, "grad_norm": 0.431640625, "learning_rate": 0.0007, "train/total_time_seconds": 259.5054631754756, "train/time_per_step_avg": 0.23832515154033898, "train/epoch_time_elapsed": 639.764641892165, "train/estimated_remaining_minutes": 5.8755953926522775}
-{"step": 1080, "epoch": 0.07280080889787664, "timestamp": 1786572738.4397485, "loss": 2.5389125823974608, "grad_norm": 0.4609375, "learning_rate": 0.0007, "train/total_time_seconds": 264.261294528842, "train/time_per_step_avg": 0.23811227962374687, "train/epoch_time_elapsed": 645.5127583183348, "train/estimated_remaining_minutes": 5.790911083811043}
-{"step": 1100, "epoch": 0.0741489720256151, "timestamp": 1786572744.230657, "loss": 2.5366193771362306, "grad_norm": 0.41796875, "learning_rate": 0.0007, "train/total_time_seconds": 269.0250007286668, "train/time_per_step_avg": 0.23822587251663208, "train/epoch_time_elapsed": 651.3036665953696, "train/estimated_remaining_minutes": 5.706590924547478}
-{"step": 1100, "epoch": 0.0741489720256151, "timestamp": 1786572759.635796, "eval_loss": 2.5286221504211426, "eval_runtime": 15.4036, "eval_samples_per_second": 618.491, "eval_steps_per_second": 0.649, "train/total_time_seconds": 269.0250007286668, "train/time_per_step_avg": 0.23822587251663208, "train/epoch_time_elapsed": 666.708803076297, "train/estimated_remaining_minutes": 5.706590924547478}
-{"step": 1120, "epoch": 0.07549713515335356, "timestamp": 1786572766.0009887, "loss": 2.501230239868164, "grad_norm": 0.42578125, "learning_rate": 0.0007, "train/total_time_seconds": 273.7952768802643, "train/time_per_step_avg": 0.23843925599008797, "train/epoch_time_elapsed": 673.0739979743958, "train/estimated_remaining_minutes": 5.6225815787911415}
-{"step": 1140, "epoch": 0.07684529828109202, "timestamp": 1786572771.7983897, "loss": 2.4896522521972657, "grad_norm": 0.419921875, "learning_rate": 0.0007, "train/total_time_seconds": 278.58514358848333, "train/time_per_step_avg": 0.2387021042406559, "train/epoch_time_elapsed": 678.8713992908597, "train/estimated_remaining_minutes": 5.53911981404002}
-{"step": 1150, "epoch": 0.07751937984496124, "timestamp": 1786572790.126436, "eval_loss": 2.5008819103240967, "eval_runtime": 15.4572, "eval_samples_per_second": 616.345, "eval_steps_per_second": 0.647, "train/total_time_seconds": 280.96107425540686, "train/time_per_step_avg": 0.2382999473810196, "train/epoch_time_elapsed": 697.1994426809251, "train/estimated_remaining_minutes": 5.497064496301438}
-{"step": 1160, "epoch": 0.07819346140883048, "timestamp": 1786572793.0218408, "loss": 2.4903682708740233, "grad_norm": 0.427734375, "learning_rate": 0.0007, "train/total_time_seconds": 283.3417424298823, "train/time_per_step_avg": 0.2383627925440669, "train/epoch_time_elapsed": 700.0948501378298, "train/estimated_remaining_minutes": 5.455142742184515}
-{"step": 1180, "epoch": 0.07954162453656892, "timestamp": 1786572798.7759607, "loss": 2.4845064163208006, "grad_norm": 0.41015625, "learning_rate": 0.0007, "train/total_time_seconds": 288.1058768518269, "train/time_per_step_avg": 0.23844582322984934, "train/epoch_time_elapsed": 705.8489697612822, "train/estimated_remaining_minutes": 5.371465500627281}
-{"step": 1200, "epoch": 0.08088978766430738, "timestamp": 1786572804.561488, "loss": 2.477877616882324, "grad_norm": 0.462890625, "learning_rate": 0.0007, "train/total_time_seconds": 292.8748792372644, "train/time_per_step_avg": 0.23849878508597613, "train/epoch_time_elapsed": 711.6344973631203, "train/estimated_remaining_minutes": 5.288018652895052}
-{"step": 1200, "epoch": 0.08088978766430738, "timestamp": 1786572820.1007786, "eval_loss": 2.466862201690674, "eval_runtime": 15.5379, "eval_samples_per_second": 613.148, "eval_steps_per_second": 0.644, "train/total_time_seconds": 292.8748792372644, "train/time_per_step_avg": 0.23849878508597613, "train/epoch_time_elapsed": 727.1737868785858, "train/estimated_remaining_minutes": 5.288018652895052}
-{"step": 1220, "epoch": 0.08223795079204584, "timestamp": 1786572826.1376908, "loss": 2.4641916275024416, "grad_norm": 0.419921875, "learning_rate": 0.0007, "train/total_time_seconds": 297.62937288731337, "train/time_per_step_avg": 0.23834096007049083, "train/epoch_time_elapsed": 733.2107002288103, "train/estimated_remaining_minutes": 5.204448050488539}
-{"step": 1240, "epoch": 0.0835861139197843, "timestamp": 1786572831.90249, "loss": 2.451630783081055, "grad_norm": 0.423828125, "learning_rate": 0.0007, "train/total_time_seconds": 302.3894924521446, "train/time_per_step_avg": 0.2380434886366129, "train/epoch_time_elapsed": 738.9754992648959, "train/estimated_remaining_minutes": 5.121112372173417}
-{"step": 1250, "epoch": 0.08426019548365352, "timestamp": 1786572850.2572594, "eval_loss": 2.44264554977417, "eval_runtime": 15.4504, "eval_samples_per_second": 616.618, "eval_steps_per_second": 0.647, "train/total_time_seconds": 304.7773395590484, "train/time_per_step_avg": 0.23816265303641557, "train/epoch_time_elapsed": 757.3302656300366, "train/estimated_remaining_minutes": 5.079622325984141}
-{"step": 1260, "epoch": 0.08493427704752275, "timestamp": 1786572853.1699786, "loss": 2.441361999511719, "grad_norm": 0.416015625, "learning_rate": 0.0007, "train/total_time_seconds": 307.1980757340789, "train/time_per_step_avg": 0.23856333304196597, "train/epoch_time_elapsed": 760.2429884597659, "train/estimated_remaining_minutes": 5.038698596696532}
-{"step": 1280, "epoch": 0.08628244017526121, "timestamp": 1786572859.062743, "loss": 2.422855186462402, "grad_norm": 0.43359375, "learning_rate": 0.0007, "train/total_time_seconds": 312.10697105154395, "train/time_per_step_avg": 0.24001094199717043, "train/epoch_time_elapsed": 766.1357527263463, "train/estimated_remaining_minutes": 4.957949279725047}
-{"step": 1300, "epoch": 0.08763060330299967, "timestamp": 1786572864.982774, "loss": 2.415398597717285, "grad_norm": 0.44140625, "learning_rate": 0.0007, "train/total_time_seconds": 317.01527301222086, "train/time_per_step_avg": 0.24140393774956465, "train/epoch_time_elapsed": 772.0557816028595, "train/estimated_remaining_minutes": 4.87715804634186}
-{"step": 1300, "epoch": 0.08763060330299967, "timestamp": 1786572880.675303, "eval_loss": 2.4175240993499756, "eval_runtime": 15.6908, "eval_samples_per_second": 607.17, "eval_steps_per_second": 0.637, "train/total_time_seconds": 317.01527301222086, "train/time_per_step_avg": 0.24140393774956465, "train/epoch_time_elapsed": 787.7483095042408, "train/estimated_remaining_minutes": 4.87715804634186}
-{"step": 1320, "epoch": 0.08897876643073811, "timestamp": 1786572886.8839746, "loss": 2.4084362030029296, "grad_norm": 0.392578125, "learning_rate": 0.0007, "train/total_time_seconds": 321.93755877017975, "train/time_per_step_avg": 0.24308185882866382, "train/epoch_time_elapsed": 793.9569843709469, "train/estimated_remaining_minutes": 4.7965444362223755}
-{"step": 1340, "epoch": 0.09032692955847657, "timestamp": 1786572892.7616043, "loss": 2.3967737197875976, "grad_norm": 0.419921875, "learning_rate": 0.0007, "train/total_time_seconds": 326.82563972473145, "train/time_per_step_avg": 0.24436147272586822, "train/epoch_time_elapsed": 799.834613699466, "train/estimated_remaining_minutes": 4.715394801998613}
-{"step": 1350, "epoch": 0.0910010111223458, "timestamp": 1786572911.1477647, "eval_loss": 2.3950557708740234, "eval_runtime": 15.451, "eval_samples_per_second": 616.594, "eval_steps_per_second": 0.647, "train/total_time_seconds": 329.26746398583055, "train/time_per_step_avg": 0.2449012442678213, "train/epoch_time_elapsed": 818.2207727767527, "train/estimated_remaining_minutes": 4.674784982514878}
-{"step": 1360, "epoch": 0.09167509268621503, "timestamp": 1786572914.119436, "loss": 2.3975946426391603, "grad_norm": 0.421875, "learning_rate": 0.0007, "train/total_time_seconds": 331.72322864085436, "train/time_per_step_avg": 0.24525152906775474, "train/epoch_time_elapsed": 821.1924460977316, "train/estimated_remaining_minutes": 4.634368635423701}
-{"step": 1380, "epoch": 0.09302325581395349, "timestamp": 1786572920.03402, "loss": 2.3763359069824217, "grad_norm": 0.408203125, "learning_rate": 0.0007, "train/total_time_seconds": 336.63740438967943, "train/time_per_step_avg": 0.24530433338135482, "train/epoch_time_elapsed": 827.1070301383734, "train/estimated_remaining_minutes": 4.5535494313579825}
-{"step": 1400, "epoch": 0.09437141894169195, "timestamp": 1786572925.9111629, "loss": 2.3692970275878906, "grad_norm": 0.404296875, "learning_rate": 0.0007, "train/total_time_seconds": 341.5246128514409, "train/time_per_step_avg": 0.24509339839220046, "train/epoch_time_elapsed": 832.9841724745929, "train/estimated_remaining_minutes": 4.472346120673631}
-{"step": 1400, "epoch": 0.09437141894169195, "timestamp": 1786572941.533417, "eval_loss": 2.3726940155029297, "eval_runtime": 15.6208, "eval_samples_per_second": 609.893, "eval_steps_per_second": 0.64, "train/total_time_seconds": 341.5246128514409, "train/time_per_step_avg": 0.24509339839220046, "train/epoch_time_elapsed": 848.6064249537885, "train/estimated_remaining_minutes": 4.472346120673631}
-{"step": 1420, "epoch": 0.0957195820694304, "timestamp": 1786572947.7788286, "loss": 2.3801626205444335, "grad_norm": 0.4375, "learning_rate": 0.0007, "train/total_time_seconds": 346.45255209878087, "train/time_per_step_avg": 0.24514993328601123, "train/epoch_time_elapsed": 854.8518381528556, "train/estimated_remaining_minutes": 4.391652068857786}
-{"step": 1440, "epoch": 0.09706774519716886, "timestamp": 1786572953.6376317, "loss": 2.3601938247680665, "grad_norm": 0.408203125, "learning_rate": 0.0007, "train/total_time_seconds": 351.3303581736982, "train/time_per_step_avg": 0.24504718448966742, "train/epoch_time_elapsed": 860.7106415070593, "train/estimated_remaining_minutes": 4.31030300537176}
-{"step": 1450, "epoch": 0.09774182676103808, "timestamp": 1786572972.0322988, "eval_loss": 2.3563716411590576, "eval_runtime": 15.4828, "eval_samples_per_second": 615.329, "eval_steps_per_second": 0.646, "train/total_time_seconds": 353.75247041508555, "train/time_per_step_avg": 0.24485006429255007, "train/epoch_time_elapsed": 879.1053058989346, "train/estimated_remaining_minutes": 4.269426367078619}
-{"step": 1460, "epoch": 0.09841590832490732, "timestamp": 1786572975.0569963, "loss": 2.3468271255493165, "grad_norm": 0.4296875, "learning_rate": 0.0007, "train/total_time_seconds": 356.239446118474, "train/time_per_step_avg": 0.24516217477619648, "train/epoch_time_elapsed": 882.1300060115755, "train/estimated_remaining_minutes": 4.229326757570925}
-{"step": 1480, "epoch": 0.09976407145264576, "timestamp": 1786572980.8820417, "loss": 2.335542678833008, "grad_norm": 0.37890625, "learning_rate": 0.0007, "train/total_time_seconds": 361.0869655944407, "train/time_per_step_avg": 0.24449561204761266, "train/epoch_time_elapsed": 887.9550512321293, "train/estimated_remaining_minutes": 4.147620550746954}
-{"step": 1500, "epoch": 0.10111223458038422, "timestamp": 1786572986.9371169, "loss": 2.3228586196899412, "grad_norm": 0.42578125, "learning_rate": 0.0007, "train/total_time_seconds": 366.1646965108812, "train/time_per_step_avg": 0.2464008365944028, "train/epoch_time_elapsed": 894.010126516223, "train/estimated_remaining_minutes": 4.06849662789868}
-{"step": 1500, "epoch": 0.10111223458038422, "timestamp": 1786573002.362648, "eval_loss": 2.3361868858337402, "eval_runtime": 15.424, "eval_samples_per_second": 617.673, "eval_steps_per_second": 0.648, "train/total_time_seconds": 366.1646965108812, "train/time_per_step_avg": 0.2464008365944028, "train/epoch_time_elapsed": 909.4356557130814, "train/estimated_remaining_minutes": 4.06849662789868}
-{"step": 1520, "epoch": 0.10246039770812268, "timestamp": 1786573008.5602589, "loss": 2.3246978759765624, "grad_norm": 0.400390625, "learning_rate": 0.0007, "train/total_time_seconds": 371.0727659277618, "train/time_per_step_avg": 0.24620213828980922, "train/epoch_time_elapsed": 915.633268546313, "train/estimated_remaining_minutes": 3.987404721592177}
-{"step": 1540, "epoch": 0.10380856083586114, "timestamp": 1786573014.4647539, "loss": 2.31418514251709, "grad_norm": 0.416015625, "learning_rate": 0.0007, "train/total_time_seconds": 375.98715218529105, "train/time_per_step_avg": 0.24656794011592864, "train/epoch_time_elapsed": 921.5377636663616, "train/estimated_remaining_minutes": 3.9063600227043227}
-{"step": 1550, "epoch": 0.10448264239973036, "timestamp": 1786573032.8496914, "eval_loss": 2.3167402744293213, "eval_runtime": 15.4168, "eval_samples_per_second": 617.963, "eval_steps_per_second": 0.649, "train/total_time_seconds": 378.4567166082561, "train/time_per_step_avg": 0.24704246193170548, "train/epoch_time_elapsed": 939.9226983599365, "train/estimated_remaining_minutes": 3.8659557072886375}
-{"step": 1560, "epoch": 0.1051567239635996, "timestamp": 1786573035.9088295, "loss": 2.3028236389160157, "grad_norm": 0.400390625, "learning_rate": 0.0007, "train/total_time_seconds": 380.98129458352923, "train/time_per_step_avg": 0.24741848465055227, "train/epoch_time_elapsed": 942.9818394444883, "train/estimated_remaining_minutes": 3.826094197740572}
-{"step": 1580, "epoch": 0.10650488709133805, "timestamp": 1786573041.8440053, "loss": 2.323251724243164, "grad_norm": 0.396484375, "learning_rate": 0.0007, "train/total_time_seconds": 385.88461896777153, "train/time_per_step_avg": 0.2479765337333083, "train/epoch_time_elapsed": 948.9170151241124, "train/estimated_remaining_minutes": 3.74487182964504}
-{"step": 1600, "epoch": 0.10785305021907651, "timestamp": 1786573047.7214646, "loss": 2.2911069869995115, "grad_norm": 0.39453125, "learning_rate": 0.0007, "train/total_time_seconds": 390.76118802279234, "train/time_per_step_avg": 0.24596491511911153, "train/epoch_time_elapsed": 954.7944736033678, "train/estimated_remaining_minutes": 3.6633861377136783}
-{"step": 1600, "epoch": 0.10785305021907651, "timestamp": 1786573063.157458, "eval_loss": 2.304603338241577, "eval_runtime": 15.4345, "eval_samples_per_second": 617.254, "eval_steps_per_second": 0.648, "train/total_time_seconds": 390.76118802279234, "train/time_per_step_avg": 0.24596491511911153, "train/epoch_time_elapsed": 970.2304658964276, "train/estimated_remaining_minutes": 3.6633861377136783}
-{"step": 1620, "epoch": 0.10920121334681497, "timestamp": 1786573069.278365, "loss": 2.288753890991211, "grad_norm": 0.396484375, "learning_rate": 0.0007, "train/total_time_seconds": 395.5747415497899, "train/time_per_step_avg": 0.24501975622028113, "train/epoch_time_elapsed": 976.3513733111322, "train/estimated_remaining_minutes": 3.581335108681225}
-{"step": 1640, "epoch": 0.11054937647455342, "timestamp": 1786573075.0791373, "loss": 2.302635383605957, "grad_norm": 0.396484375, "learning_rate": 0.0007, "train/total_time_seconds": 400.3638037070632, "train/time_per_step_avg": 0.24376651521772147, "train/epoch_time_elapsed": 982.152145486325, "train/estimated_remaining_minutes": 3.499114544594251}
-{"step": 1650, "epoch": 0.11122345803842265, "timestamp": 1786573093.4526207, "eval_loss": 2.2875566482543945, "eval_runtime": 15.4822, "eval_samples_per_second": 615.351, "eval_steps_per_second": 0.646, "train/total_time_seconds": 402.75052274018526, "train/time_per_step_avg": 0.2429380613192916, "train/epoch_time_elapsed": 1000.5256290659308, "train/estimated_remaining_minutes": 3.457959033627853}
-{"step": 1660, "epoch": 0.11189753960229187, "timestamp": 1786573096.3818655, "loss": 2.2855497360229493, "grad_norm": 0.38671875, "learning_rate": 0.0007, "train/total_time_seconds": 405.1449579000473, "train/time_per_step_avg": 0.2416366331651807, "train/epoch_time_elapsed": 1003.4548746943474, "train/estimated_remaining_minutes": 3.4168851871088326}
-{"step": 1680, "epoch": 0.11324570273003033, "timestamp": 1786573102.5616276, "loss": 2.26898307800293, "grad_norm": 0.39453125, "learning_rate": 0.0007, "train/total_time_seconds": 410.2858841829002, "train/time_per_step_avg": 0.2440126521512866, "train/epoch_time_elapsed": 1009.634637080133, "train/estimated_remaining_minutes": 3.337643105456132}
-{"step": 1700, "epoch": 0.11459386585776879, "timestamp": 1786573108.5055392, "loss": 2.278878021240234, "grad_norm": 0.40625, "learning_rate": 0.0007, "train/total_time_seconds": 415.16579672694206, "train/time_per_step_avg": 0.24404608704149722, "train/epoch_time_elapsed": 1015.5785482823849, "train/estimated_remaining_minutes": 3.2562023272701337}
-{"step": 1700, "epoch": 0.11459386585776879, "timestamp": 1786573124.0211952, "eval_loss": 2.272601366043091, "eval_runtime": 15.5142, "eval_samples_per_second": 614.084, "eval_steps_per_second": 0.645, "train/total_time_seconds": 415.16579672694206, "train/time_per_step_avg": 0.24404608704149722, "train/epoch_time_elapsed": 1031.0942024588585, "train/estimated_remaining_minutes": 3.2562023272701337}
-{"step": 1720, "epoch": 0.11594202898550725, "timestamp": 1786573130.173951, "loss": 2.25760498046875, "grad_norm": 0.51953125, "learning_rate": 0.0007, "train/total_time_seconds": 419.9892845861614, "train/time_per_step_avg": 0.2441454303637147, "train/epoch_time_elapsed": 1037.2469600513577, "train/estimated_remaining_minutes": 3.1743376160581964}
-{"step": 1740, "epoch": 0.1172901921132457, "timestamp": 1786573135.9923048, "loss": 2.2478647232055664, "grad_norm": 0.41015625, "learning_rate": 0.0007, "train/total_time_seconds": 424.7985208891332, "train/time_per_step_avg": 0.24434717182070018, "train/epoch_time_elapsed": 1043.06531425938, "train/estimated_remaining_minutes": 3.0924030256297055}
-{"step": 1750, "epoch": 0.11796427367711493, "timestamp": 1786573154.3798609, "eval_loss": 2.2599399089813232, "eval_runtime": 15.5111, "eval_samples_per_second": 614.204, "eval_steps_per_second": 0.645, "train/total_time_seconds": 427.1772727891803, "train/time_per_step_avg": 0.2442675004899502, "train/epoch_time_elapsed": 1061.4528689235449, "train/estimated_remaining_minutes": 3.051266234208431}
-{"step": 1760, "epoch": 0.11863835524098416, "timestamp": 1786573157.268258, "loss": 2.253020668029785, "grad_norm": 0.4140625, "learning_rate": 0.0007, "train/total_time_seconds": 429.5491810142994, "train/time_per_step_avg": 0.2440422311425209, "train/epoch_time_elapsed": 1064.3412677310407, "train/estimated_remaining_minutes": 3.0100984275623253}
-{"step": 1780, "epoch": 0.11998651836872262, "timestamp": 1786573163.0509636, "loss": 2.2667808532714844, "grad_norm": 0.40625, "learning_rate": 0.0007, "train/total_time_seconds": 434.329084828496, "train/time_per_step_avg": 0.2404320064559579, "train/epoch_time_elapsed": 1070.1239728368819, "train/estimated_remaining_minutes": 2.928061246034804}
-{"step": 1800, "epoch": 0.12133468149646107, "timestamp": 1786573168.7891948, "loss": 2.2645755767822267, "grad_norm": 0.38671875, "learning_rate": 0.0007, "train/total_time_seconds": 439.0793381296098, "train/time_per_step_avg": 0.2391354140266776, "train/epoch_time_elapsed": 1075.8622042126954, "train/estimated_remaining_minutes": 2.8458845989882118}
-{"step": 1800, "epoch": 0.12133468149646107, "timestamp": 1786573184.3052278, "eval_loss": 2.2453370094299316, "eval_runtime": 15.5145, "eval_samples_per_second": 614.069, "eval_steps_per_second": 0.645, "train/total_time_seconds": 439.0793381296098, "train/time_per_step_avg": 0.2391354140266776, "train/epoch_time_elapsed": 1091.3782351277769, "train/estimated_remaining_minutes": 2.8458845989882118}
-{"step": 1820, "epoch": 0.12268284462419952, "timestamp": 1786573190.3325443, "loss": 2.2548261642456056, "grad_norm": 0.40625, "learning_rate": 0.0007, "train/total_time_seconds": 443.8459137752652, "train/time_per_step_avg": 0.23856629189103842, "train/epoch_time_elapsed": 1097.4055539146066, "train/estimated_remaining_minutes": 2.763875653545607}
-{"step": 1840, "epoch": 0.12403100775193798, "timestamp": 1786573196.231179, "loss": 2.2223764419555665, "grad_norm": 0.384765625, "learning_rate": 0.0007, "train/total_time_seconds": 448.66495551913977, "train/time_per_step_avg": 0.23866434630006553, "train/epoch_time_elapsed": 1103.3041876107454, "train/estimated_remaining_minutes": 2.6822361471252925}
-{"step": 1850, "epoch": 0.12470508931580722, "timestamp": 1786573214.5403116, "eval_loss": 2.23197078704834, "eval_runtime": 15.4389, "eval_samples_per_second": 617.079, "eval_steps_per_second": 0.648, "train/total_time_seconds": 451.04221741110086, "train/time_per_step_avg": 0.23864944621920586, "train/epoch_time_elapsed": 1121.613317873329, "train/estimated_remaining_minutes": 2.641238210065005}
-{"step": 1860, "epoch": 0.12537917087967643, "timestamp": 1786573217.4205904, "loss": 2.2210620880126952, "grad_norm": 0.388671875, "learning_rate": 0.0007, "train/total_time_seconds": 453.42419058829546, "train/time_per_step_avg": 0.23875009573996067, "train/epoch_time_elapsed": 1124.4935989007354, "train/estimated_remaining_minutes": 2.600282096563702}
-{"step": 1880, "epoch": 0.1267273340074149, "timestamp": 1786573223.5239706, "loss": 2.2141876220703125, "grad_norm": 0.373046875, "learning_rate": 0.0007, "train/total_time_seconds": 458.4944286607206, "train/time_per_step_avg": 0.24165343832224606, "train/epoch_time_elapsed": 1130.596980381757, "train/estimated_remaining_minutes": 2.520093490865663}
-{"step": 1900, "epoch": 0.12807549713515334, "timestamp": 1786573229.308081, "loss": 2.221418762207031, "grad_norm": 0.396484375, "learning_rate": 0.0007, "train/total_time_seconds": 463.28024032339454, "train/time_per_step_avg": 0.24200902193784712, "train/epoch_time_elapsed": 1136.3810905963182, "train/estimated_remaining_minutes": 2.438317054333656}
-{"step": 1900, "epoch": 0.12807549713515334, "timestamp": 1786573244.7756603, "eval_loss": 2.221304416656494, "eval_runtime": 15.4661, "eval_samples_per_second": 615.991, "eval_steps_per_second": 0.647, "train/total_time_seconds": 463.28024032339454, "train/time_per_step_avg": 0.24200902193784712, "train/epoch_time_elapsed": 1151.848667576909, "train/estimated_remaining_minutes": 2.438317054333656}
-{"step": 1920, "epoch": 0.12942366026289182, "timestamp": 1786573250.8332932, "loss": 2.199698257446289, "grad_norm": 0.37890625, "learning_rate": 0.0007, "train/total_time_seconds": 468.0679714716971, "train/time_per_step_avg": 0.24222057696431876, "train/epoch_time_elapsed": 1157.9063024520874, "train/estimated_remaining_minutes": 2.356592217479031}
-{"step": 1940, "epoch": 0.13077182339063026, "timestamp": 1786573256.7804224, "loss": 2.210783004760742, "grad_norm": 0.419921875, "learning_rate": 0.0007, "train/total_time_seconds": 472.93661507219076, "train/time_per_step_avg": 0.24271659553050995, "train/epoch_time_elapsed": 1163.8534313961864, "train/estimated_remaining_minutes": 2.275296429900574}
-{"step": 1950, "epoch": 0.13144590495449948, "timestamp": 1786573275.0893812, "eval_loss": 2.2097606658935547, "eval_runtime": 15.443, "eval_samples_per_second": 616.913, "eval_steps_per_second": 0.648, "train/total_time_seconds": 475.3099448233843, "train/time_per_step_avg": 0.2426772741228342, "train/epoch_time_elapsed": 1182.162388883531, "train/estimated_remaining_minutes": 2.234362988485995}
-{"step": 1960, "epoch": 0.13211998651836873, "timestamp": 1786573277.9565737, "loss": 2.1856407165527343, "grad_norm": 0.392578125, "learning_rate": 0.0007, "train/total_time_seconds": 477.6850075908005, "train/time_per_step_avg": 0.24260817002505064, "train/epoch_time_elapsed": 1185.0295825749636, "train/estimated_remaining_minutes": 2.1934515654679614}
-{"step": 1980, "epoch": 0.13346814964610718, "timestamp": 1786573283.7114468, "loss": 2.2084430694580077, "grad_norm": 0.404296875, "learning_rate": 0.0007, "train/total_time_seconds": 482.44906667247415, "train/time_per_step_avg": 0.2395463801175356, "train/epoch_time_elapsed": 1190.7844559438527, "train/estimated_remaining_minutes": 2.111729921462008}
-{"step": 2000, "epoch": 0.13481631277384565, "timestamp": 1786573289.5163822, "loss": 2.181157112121582, "grad_norm": 0.376953125, "learning_rate": 0.0007, "train/total_time_seconds": 487.2355602942407, "train/time_per_step_avg": 0.23955319970846176, "train/epoch_time_elapsed": 1196.5893914178014, "train/estimated_remaining_minutes": 2.0301481678926696}
-{"step": 2000, "epoch": 0.13481631277384565, "timestamp": 1786573305.0540488, "eval_loss": 2.2002651691436768, "eval_runtime": 15.5362, "eval_samples_per_second": 613.213, "eval_steps_per_second": 0.644, "train/total_time_seconds": 487.2355602942407, "train/time_per_step_avg": 0.23955319970846176, "train/epoch_time_elapsed": 1212.1270550377667, "train/estimated_remaining_minutes": 2.0301481678926696}
-{"step": 2020, "epoch": 0.1361644759015841, "timestamp": 1786573311.1279309, "loss": 2.1924610137939453, "grad_norm": 0.404296875, "learning_rate": 0.0007, "train/total_time_seconds": 491.9965412430465, "train/time_per_step_avg": 0.2392856977134943, "train/epoch_time_elapsed": 1218.200940232724, "train/estimated_remaining_minutes": 1.948501153437808}
-{"step": 2040, "epoch": 0.13751263902932254, "timestamp": 1786573316.9711125, "loss": 2.173542594909668, "grad_norm": 0.4140625, "learning_rate": 0.0007, "train/total_time_seconds": 496.84305188432336, "train/time_per_step_avg": 0.23906436812132598, "train/epoch_time_elapsed": 1224.0441219843924, "train/estimated_remaining_minutes": 1.8672206198267056}
-{"step": 2050, "epoch": 0.1381867205931918, "timestamp": 1786573335.427487, "eval_loss": 2.188297986984253, "eval_runtime": 15.5479, "eval_samples_per_second": 612.751, "eval_steps_per_second": 0.643, "train/total_time_seconds": 499.23411824181676, "train/time_per_step_avg": 0.23924173418432473, "train/epoch_time_elapsed": 1242.5004936791956, "train/estimated_remaining_minutes": 1.8264662862505492}
-{"step": 2060, "epoch": 0.138860802157061, "timestamp": 1786573338.5883698, "loss": 2.192971038818359, "grad_norm": 0.39453125, "learning_rate": 0.0007, "train/total_time_seconds": 501.9024883285165, "train/time_per_step_avg": 0.2421748073771596, "train/epoch_time_elapsed": 1245.661379236728, "train/estimated_remaining_minutes": 1.7867078872536186}
-{"step": 2080, "epoch": 0.14020896528479945, "timestamp": 1786573344.3097484, "loss": 2.168643569946289, "grad_norm": 0.388671875, "learning_rate": 0.0007, "train/total_time_seconds": 506.6452153734863, "train/time_per_step_avg": 0.24196148701012135, "train/epoch_time_elapsed": 1251.382757332176, "train/estimated_remaining_minutes": 1.7050560132761556}
-{"step": 2100, "epoch": 0.14155712841253792, "timestamp": 1786573350.2870846, "loss": 2.1587839126586914, "grad_norm": 0.37109375, "learning_rate": 0.0007, "train/total_time_seconds": 511.5303033441305, "train/time_per_step_avg": 0.24294743049889803, "train/epoch_time_elapsed": 1257.360093627125, "train/estimated_remaining_minutes": 1.6239057249020017}
-{"step": 2100, "epoch": 0.14155712841253792, "timestamp": 1786573365.7820985, "eval_loss": 2.179719924926758, "eval_runtime": 15.4936, "eval_samples_per_second": 614.899, "eval_steps_per_second": 0.645, "train/total_time_seconds": 511.5303033441305, "train/time_per_step_avg": 0.24294743049889803, "train/epoch_time_elapsed": 1272.8551050648093, "train/estimated_remaining_minutes": 1.6239057249020017}
-{"step": 2120, "epoch": 0.14290529154027637, "timestamp": 1786573371.7915163, "loss": 2.1700040817260744, "grad_norm": 0.384765625, "learning_rate": 0.0007, "train/total_time_seconds": 516.2742769606411, "train/time_per_step_avg": 0.24277735717594623, "train/epoch_time_elapsed": 1278.8645255304873, "train/estimated_remaining_minutes": 1.5423288148195256}
-{"step": 2140, "epoch": 0.14425345466801484, "timestamp": 1786573377.572285, "loss": 2.1694923400878907, "grad_norm": 0.388671875, "learning_rate": 0.0007, "train/total_time_seconds": 521.0364815220237, "train/time_per_step_avg": 0.2419342963770032, "train/epoch_time_elapsed": 1284.6452944390476, "train/estimated_remaining_minutes": 1.4608499481925898}
-{"step": 2150, "epoch": 0.14492753623188406, "timestamp": 1786573396.1551394, "eval_loss": 2.1717145442962646, "eval_runtime": 15.5797, "eval_samples_per_second": 611.5, "eval_steps_per_second": 0.642, "train/total_time_seconds": 523.5091845020652, "train/time_per_step_avg": 0.24275066260248424, "train/epoch_time_elapsed": 1303.22814546898, "train/estimated_remaining_minutes": 1.420373756400952}
-{"step": 2160, "epoch": 0.14560161779575329, "timestamp": 1786573399.0292964, "loss": 2.1554704666137696, "grad_norm": 0.3828125, "learning_rate": 0.0007, "train/total_time_seconds": 525.887499935925, "train/time_per_step_avg": 0.23985011607408524, "train/epoch_time_elapsed": 1306.1023048982024, "train/estimated_remaining_minutes": 1.3796431325479512}
-{"step": 2180, "epoch": 0.14694978092349173, "timestamp": 1786573404.761281, "loss": 2.1634592056274413, "grad_norm": 0.388671875, "learning_rate": 0.0007, "train/total_time_seconds": 530.633124396205, "train/time_per_step_avg": 0.23987909022718668, "train/epoch_time_elapsed": 1311.8342905528843, "train/estimated_remaining_minutes": 1.2981850138133455}
-{"step": 2200, "epoch": 0.1482979440512302, "timestamp": 1786573410.5145934, "loss": 2.139862632751465, "grad_norm": 0.3828125, "learning_rate": 0.0007, "train/total_time_seconds": 535.4025480635464, "train/time_per_step_avg": 0.23872244719415903, "train/epoch_time_elapsed": 1317.5876028500497, "train/estimated_remaining_minutes": 1.2168239728716965}
-{"step": 2200, "epoch": 0.1482979440512302, "timestamp": 1786573426.0522175, "eval_loss": 2.1605989933013916, "eval_runtime": 15.5361, "eval_samples_per_second": 613.218, "eval_steps_per_second": 0.644, "train/total_time_seconds": 535.4025480635464, "train/time_per_step_avg": 0.23872244719415903, "train/epoch_time_elapsed": 1333.1252236925066, "train/estimated_remaining_minutes": 1.2168239728716965}
-{"step": 2220, "epoch": 0.14964610717896865, "timestamp": 1786573432.105549, "loss": 2.14636173248291, "grad_norm": 0.376953125, "learning_rate": 0.0007, "train/total_time_seconds": 540.1536929272115, "train/time_per_step_avg": 0.23879415966570378, "train/epoch_time_elapsed": 1339.178558100015, "train/estimated_remaining_minutes": 1.1354582133605047}
-{"step": 2240, "epoch": 0.15099427030670712, "timestamp": 1786573437.840799, "loss": 2.1564199447631838, "grad_norm": 0.390625, "learning_rate": 0.0007, "train/total_time_seconds": 544.9023728482425, "train/time_per_step_avg": 0.23865891326218844, "train/epoch_time_elapsed": 1344.913807529956, "train/estimated_remaining_minutes": 1.0541266141409453}
-{"step": 2250, "epoch": 0.15166835187057634, "timestamp": 1786573456.2842257, "eval_loss": 2.15016770362854, "eval_runtime": 15.5547, "eval_samples_per_second": 612.485, "eval_steps_per_second": 0.643, "train/total_time_seconds": 547.29029269889, "train/time_per_step_avg": 0.23781108196824788, "train/epoch_time_elapsed": 1363.357232388109, "train/estimated_remaining_minutes": 1.0135005420349814}
-{"step": 2260, "epoch": 0.15234243343444556, "timestamp": 1786573459.473562, "loss": 2.13549747467041, "grad_norm": 0.396484375, "learning_rate": 0.0007, "train/total_time_seconds": 549.9809849858284, "train/time_per_step_avg": 0.24093485049903393, "train/epoch_time_elapsed": 1366.5465720668435, "train/estimated_remaining_minutes": 0.9734176725412892}
-{"step": 2280, "epoch": 0.15369059656218403, "timestamp": 1786573465.3814912, "loss": 2.1465373992919923, "grad_norm": 0.3671875, "learning_rate": 0.0007, "train/total_time_seconds": 554.8979406356812, "train/time_per_step_avg": 0.24264816239476203, "train/epoch_time_elapsed": 1372.4545010887086, "train/estimated_remaining_minutes": 0.8923797290924697}
-{"step": 2300, "epoch": 0.15503875968992248, "timestamp": 1786573471.262166, "loss": 2.1275094985961913, "grad_norm": 0.376953125, "learning_rate": 0.0007, "train/total_time_seconds": 559.7959095165133, "train/time_per_step_avg": 0.24393361452966927, "train/epoch_time_elapsed": 1378.3351758122444, "train/estimated_remaining_minutes": 0.8112984195891497}
-{"step": 2300, "epoch": 0.15503875968992248, "timestamp": 1786573486.7685223, "eval_loss": 2.1419694423675537, "eval_runtime": 15.5048, "eval_samples_per_second": 614.454, "eval_steps_per_second": 0.645, "train/total_time_seconds": 559.7959095165133, "train/time_per_step_avg": 0.24393361452966927, "train/epoch_time_elapsed": 1393.841528620571, "train/estimated_remaining_minutes": 0.8112984195891497}
-{"step": 2320, "epoch": 0.15638692281766095, "timestamp": 1786573493.018277, "loss": 2.137742042541504, "grad_norm": 0.404296875, "learning_rate": 0.0007, "train/total_time_seconds": 564.705421499908, "train/time_per_step_avg": 0.24551728572696446, "train/epoch_time_elapsed": 1400.0912870690227, "train/estimated_remaining_minutes": 0.730222527801605}
-{"step": 2340, "epoch": 0.1577350859453994, "timestamp": 1786573498.9125712, "loss": 2.128017234802246, "grad_norm": 0.369140625, "learning_rate": 0.0007, "train/total_time_seconds": 569.6138756908476, "train/time_per_step_avg": 0.24711502842605115, "train/epoch_time_elapsed": 1405.9855805039406, "train/estimated_remaining_minutes": 0.6491326218699118}
-{"step": 2350, "epoch": 0.15840916750926862, "timestamp": 1786573517.4188173, "eval_loss": 2.1339120864868164, "eval_runtime": 15.5839, "eval_samples_per_second": 611.337, "eval_steps_per_second": 0.642, "train/total_time_seconds": 572.0436832271516, "train/time_per_step_avg": 0.24753390528261662, "train/epoch_time_elapsed": 1424.4918251223862, "train/estimated_remaining_minutes": 0.6085571098161188}
-{"step": 2360, "epoch": 0.15908324907313784, "timestamp": 1786573520.4341505, "loss": 2.130332374572754, "grad_norm": 0.376953125, "learning_rate": 0.0007, "train/total_time_seconds": 574.5273701362312, "train/time_per_step_avg": 0.24546385150402783, "train/epoch_time_elapsed": 1427.5071605332196, "train/estimated_remaining_minutes": 0.5680355354454263}
-{"step": 2380, "epoch": 0.1604314122008763, "timestamp": 1786573526.3541203, "loss": 2.1314908981323244, "grad_norm": 0.384765625, "learning_rate": 0.0007, "train/total_time_seconds": 579.4237385764718, "train/time_per_step_avg": 0.24525797940790653, "train/epoch_time_elapsed": 1433.427129920572, "train/estimated_remaining_minutes": 0.4869107046861108}
-{"step": 2400, "epoch": 0.16177957532861476, "timestamp": 1786573532.227129, "loss": 2.112477111816406, "grad_norm": 0.380859375, "learning_rate": 0.0007, "train/total_time_seconds": 584.3077703565359, "train/time_per_step_avg": 0.24511860840022565, "train/epoch_time_elapsed": 1439.3001387827098, "train/estimated_remaining_minutes": 0.4057692849698166}
-{"step": 2400, "epoch": 0.16177957532861476, "timestamp": 1786573547.7280645, "eval_loss": 2.1304807662963867, "eval_runtime": 15.4992, "eval_samples_per_second": 614.676, "eval_steps_per_second": 0.645, "train/total_time_seconds": 584.3077703565359, "train/time_per_step_avg": 0.24511860840022565, "train/epoch_time_elapsed": 1454.8010725155473, "train/estimated_remaining_minutes": 0.4057692849698166}
-{"step": 2420, "epoch": 0.16312773845635323, "timestamp": 1786573553.9683285, "loss": 2.1156797409057617, "grad_norm": 0.396484375, "learning_rate": 0.0007, "train/total_time_seconds": 589.2346196584404, "train/time_per_step_avg": 0.2452919815853238, "train/epoch_time_elapsed": 1461.0413386188447, "train/estimated_remaining_minutes": 0.3246471733655319}
-{"step": 2440, "epoch": 0.16447590158409167, "timestamp": 1786573559.86106, "loss": 2.104845428466797, "grad_norm": 0.37890625, "learning_rate": 0.0007, "train/total_time_seconds": 594.1228236779571, "train/time_per_step_avg": 0.24508947987109422, "train/epoch_time_elapsed": 1466.934069506824, "train/estimated_remaining_minutes": 0.24349296052375288}
-{"step": 2450, "epoch": 0.1651499831479609, "timestamp": 1786573578.2964396, "eval_loss": 2.1203832626342773, "eval_runtime": 15.4857, "eval_samples_per_second": 615.212, "eval_steps_per_second": 0.646, "train/total_time_seconds": 596.5711809359491, "train/time_per_step_avg": 0.24527497708797455, "train/epoch_time_elapsed": 1485.369447492063, "train/estimated_remaining_minutes": 0.2029153676652888}
-{"step": 2460, "epoch": 0.16582406471183014, "timestamp": 1786573581.4497626, "loss": 2.1162158966064455, "grad_norm": 0.3828125, "learning_rate": 0.0007, "train/total_time_seconds": 599.2305207550526, "train/time_per_step_avg": 0.24703150618821382, "train/epoch_time_elapsed": 1488.5227723307908, "train/estimated_remaining_minutes": 0.16239309505556979}
-{"step": 2480, "epoch": 0.1671722278395686, "timestamp": 1786573587.4402347, "loss": 2.116878890991211, "grad_norm": 0.3828125, "learning_rate": 0.0007, "train/total_time_seconds": 604.1884610690176, "train/time_per_step_avg": 0.24764722492545843, "train/epoch_time_elapsed": 1494.5132445953786, "train/estimated_remaining_minutes": 0.08120812648777118}
-{"step": 2500, "epoch": 0.16852039096730703, "timestamp": 1786573593.28358, "loss": 2.1019779205322267, "grad_norm": 0.3828125, "learning_rate": 0.0007, "train/total_time_seconds": 609.0446489229798, "train/time_per_step_avg": 0.2473687856644392, "train/epoch_time_elapsed": 1500.356590077281, "train/estimated_remaining_minutes": 0.0}
-{"step": 2500, "epoch": 0.16852039096730703, "timestamp": 1786573608.740347, "eval_loss": 2.1118547916412354, "eval_runtime": 15.4552, "eval_samples_per_second": 616.427, "eval_steps_per_second": 0.647, "train/total_time_seconds": 609.0446489229798, "train/time_per_step_avg": 0.2473687856644392, "train/epoch_time_elapsed": 1515.8133544363081, "train/estimated_remaining_minutes": 0.0}
-{"step": 2500, "epoch": 0.16852039096730703, "timestamp": 1786573609.0443015, "train_runtime": 1516.872, "train_samples_per_second": 105.48, "train_steps_per_second": 1.648, "total_flos": 8065709506560000.0, "train_loss": 3.0034169891357423, "train/total_time_seconds": 609.0446489229798, "train/time_per_step_avg": 0.2473687856644392, "train/epoch_time_elapsed": 1516.1173098236322, "train/estimated_remaining_minutes": 0.0}
-{"step": 2500, "epoch": 0.16852039096730703, "timestamp": 1786573624.5945957, "eval_loss": 2.1118547916412354, "eval_runtime": 15.5468, "eval_samples_per_second": 612.794, "eval_steps_per_second": 0.643, "train/total_time_seconds": 609.0446489229798, "train/time_per_step_avg": 0.2473687856644392, "train/epoch_time_elapsed": 1531.6676030904055, "train/estimated_remaining_minutes": 0.0}
+{"step": 20, "epoch": 0.0013481631277384564, "timestamp": 1786738896.6145349, "loss": 8.29943389892578, "grad_norm": 2.125, "learning_rate": 2.66e-05, "train/total_time_seconds": 6.765735510736704, "train/time_per_step_avg": 0.33828677553683517, "train/epoch_time_elapsed": 7.842617053538561, "train/estimated_remaining_minutes": 13.98252005552252}
+{"step": 40, "epoch": 0.002696326255476913, "timestamp": 1786738904.1699708, "loss": 8.03638916015625, "grad_norm": 1.3046875, "learning_rate": 5.46e-05, "train/total_time_seconds": 13.31200486049056, "train/time_per_step_avg": 0.332800121512264, "train/epoch_time_elapsed": 15.398053739219904, "train/estimated_remaining_minutes": 13.644804982002825}
+{"step": 60, "epoch": 0.004044489383215369, "timestamp": 1786738911.668541, "loss": 7.834577941894532, "grad_norm": 1.2734375, "learning_rate": 8.259999999999999e-05, "train/total_time_seconds": 19.815627731382847, "train/time_per_step_avg": 0.3302604621897141, "train/epoch_time_elapsed": 22.89662381634116, "train/estimated_remaining_minutes": 13.430592129048373}
+{"step": 80, "epoch": 0.005392652510953826, "timestamp": 1786738919.0760295, "loss": 7.5719154357910154, "grad_norm": 1.265625, "learning_rate": 0.0001106, "train/total_time_seconds": 26.208368558436632, "train/time_per_step_avg": 0.3276046069804579, "train/epoch_time_elapsed": 30.30411148071289, "train/estimated_remaining_minutes": 13.21338581487847}
+{"step": 100, "epoch": 0.006740815638692282, "timestamp": 1786738926.4269438, "loss": 7.23726806640625, "grad_norm": 1.1953125, "learning_rate": 0.0001386, "train/total_time_seconds": 32.56988685950637, "train/time_per_step_avg": 0.3256988685950637, "train/epoch_time_elapsed": 37.65502642467618, "train/estimated_remaining_minutes": 13.027954743802548}
+{"step": 120, "epoch": 0.008088978766430738, "timestamp": 1786738934.243374, "loss": 6.864923095703125, "grad_norm": 1.1484375, "learning_rate": 0.00016659999999999998, "train/total_time_seconds": 39.04843807592988, "train/time_per_step_avg": 0.3228270256519318, "train/epoch_time_elapsed": 45.471456445753574, "train/estimated_remaining_minutes": 12.90767814176571}
+{"step": 140, "epoch": 0.009437141894169195, "timestamp": 1786738941.632826, "loss": 6.509650421142578, "grad_norm": 1.0234375, "learning_rate": 0.00019460000000000001, "train/total_time_seconds": 45.43964921310544, "train/time_per_step_avg": 0.3212764435261488, "train/epoch_time_elapsed": 52.86090863123536, "train/estimated_remaining_minutes": 12.766377636062957}
+{"step": 160, "epoch": 0.010785305021907651, "timestamp": 1786738949.1611085, "loss": 6.204092788696289, "grad_norm": 1.03125, "learning_rate": 0.0002226, "train/total_time_seconds": 51.89764886721969, "train/time_per_step_avg": 0.3208202113583684, "train/epoch_time_elapsed": 60.38919149339199, "train/estimated_remaining_minutes": 12.650051911384798}
+{"step": 180, "epoch": 0.012133468149646108, "timestamp": 1786738956.5629447, "loss": 5.9408447265625, "grad_norm": 0.87109375, "learning_rate": 0.00025059999999999997, "train/total_time_seconds": 58.30260116606951, "train/time_per_step_avg": 0.3209423260763288, "train/epoch_time_elapsed": 67.79102772846818, "train/estimated_remaining_minutes": 12.524262472711229}
+{"step": 200, "epoch": 0.013481631277384564, "timestamp": 1786738964.115336, "loss": 5.677399826049805, "grad_norm": 0.76171875, "learning_rate": 0.0002786, "train/total_time_seconds": 64.86148905009031, "train/time_per_step_avg": 0.32291602190583946, "train/epoch_time_elapsed": 75.34341878071427, "train/estimated_remaining_minutes": 12.43178540126731}
+{"step": 220, "epoch": 0.01482979440512302, "timestamp": 1786738971.9829245, "loss": 5.419148254394531, "grad_norm": 0.68359375, "learning_rate": 0.00030659999999999997, "train/total_time_seconds": 71.42476844415069, "train/time_per_step_avg": 0.32376330368220807, "train/epoch_time_elapsed": 83.2110064253211, "train/estimated_remaining_minutes": 12.337005458535119}
+{"step": 240, "epoch": 0.016177957532861477, "timestamp": 1786738979.4075286, "loss": 5.138995742797851, "grad_norm": 0.65625, "learning_rate": 0.0003346, "train/total_time_seconds": 77.84708186611533, "train/time_per_step_avg": 0.32407432653009893, "train/epoch_time_elapsed": 90.63561167940497, "train/estimated_remaining_minutes": 12.217667015098657}
+{"step": 260, "epoch": 0.01752612066059993, "timestamp": 1786738987.04185, "loss": 4.88677749633789, "grad_norm": 1.265625, "learning_rate": 0.00036260000000000003, "train/total_time_seconds": 84.47356577962637, "train/time_per_step_avg": 0.3257591691240668, "train/epoch_time_elapsed": 98.26993289217353, "train/estimated_remaining_minutes": 12.129537650407888}

zain/Activation/wandb/debug-internal.log

@@ -1,23 +1,21 @@
-{"time":"2026-08-14T20:16:15.759688663Z","level":"INFO","msg":"wandb-core"}
-{"time":"2026-08-14T20:16:15.75981584Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
-{"time":"2026-08-14T20:16:16.019892925Z","level":"INFO","msg":"stream: created new stream","id":"w8mr4w6k"}
-{"time":"2026-08-14T20:16:16.01997882Z","level":"INFO","msg":"handler: started"}
-{"time":"2026-08-14T20:16:16.020075043Z","level":"INFO","msg":"stream: started"}
-{"time":"2026-08-14T20:16:16.02008577Z","level":"INFO","msg":"writer: started","stream_id":"w8mr4w6k"}
-{"time":"2026-08-14T20:16:16.020113292Z","level":"INFO","msg":"sender: started"}
-{"time":"2026-08-14T20:16:16.531359023Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
-{"time":"2026-08-14T20:16:16.686067824Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-14T20:16:31.532246567Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":2,"uploaded_len":2}
-{"time":"2026-08-14T20:16:31.65014999Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-14T20:16:46.53154926Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-14T20:16:46.697907798Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-14T20:17:01.532416577Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-14T20:17:01.661940731Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-14T20:17:16.532407844Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":5,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-14T20:17:16.655117007Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-14T20:17:31.531597694Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":7,"events_lines":2,"console_offset":0,"console_lines":1}
-{"time":"2026-08-14T20:17:31.671651312Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-14T20:17:46.532231919Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":9,"events_lines":2,"console_offset":2,"console_lines":10}
-{"time":"2026-08-14T20:17:46.65675743Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
-{"time":"2026-08-14T20:18:01.532320003Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":11,"events_lines":2,"console_offset":10,"console_lines":1}
-{"time":"2026-08-14T20:18:01.703907556Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:28.155354055Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T20:21:28.155495403Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T20:21:28.41445694Z","level":"INFO","msg":"stream: created new stream","id":"twgo1e89"}
+{"time":"2026-08-14T20:21:28.41452958Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T20:21:28.414611994Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T20:21:28.414621977Z","level":"INFO","msg":"writer: started","stream_id":"twgo1e89"}
+{"time":"2026-08-14T20:21:28.414641036Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T20:21:28.77231947Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:21:28.885103894Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:43.772746583Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":2,"uploaded_len":2}
+{"time":"2026-08-14T20:21:43.910826986Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:58.773175035Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":2,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:21:58.957050518Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:13.773475474Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":2,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:22:13.977519477Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:28.773095938Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":2,"events_offset":5,"events_lines":2,"console_offset":2,"console_lines":11}
+{"time":"2026-08-14T20:22:28.955788633Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:43.773237697Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":2,"events_offset":7,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:22:43.926628755Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:58.772568445Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":2,"events_offset":9,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:22:58.931284477Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/debug.log

@@ -1,19 +1,20 @@
-2026-08-14 20:16:15,758 INFO    MainThread:409188 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_201615-w8mr4w6k/logs/debug.log
-2026-08-14 20:16:15,758 INFO    MainThread:409188 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_201615-w8mr4w6k/logs/debug-internal.log
-2026-08-14 20:16:15,758 INFO    MainThread:409188 [wandb_init.py:init():772] calling init triggers
-2026-08-14 20:16:15,758 INFO    MainThread:409188 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+2026-08-14 20:21:28,153 INFO    MainThread:171901 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_202128-twgo1e89/logs/debug.log
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_202128-twgo1e89/logs/debug-internal.log
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():772] calling init triggers
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
 config: {'_wandb': {}}
-2026-08-14 20:16:15,758 INFO    MainThread:409188 [wandb_init.py:init():820] starting backend
-2026-08-14 20:16:15,758 INFO    MainThread:409188 [wandb_init.py:init():835] sending inform_init request
-2026-08-14 20:16:16,020 INFO    MainThread:409188 [wandb_init.py:init():840] backend started and connected
-2026-08-14 20:16:16,021 INFO    MainThread:409188 [wandb_init.py:init():910] updated telemetry
-2026-08-14 20:16:16,028 INFO    MainThread:409188 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
-2026-08-14 20:16:16,449 INFO    MainThread:409188 [wandb_init.py:init():978] starting run threads in backend
-2026-08-14 20:16:16,526 INFO    MainThread:409188 [wandb_run.py:_console_start():2621] atexit reg
-2026-08-14 20:16:16,526 INFO    MainThread:409188 [wandb_run.py:_redirect():2471] redirect: wrap_raw
-2026-08-14 20:16:16,526 INFO    MainThread:409188 [wandb_run.py:_redirect():2540] Wrapping output streams.
-2026-08-14 20:16:16,526 INFO    MainThread:409188 [wandb_run.py:_redirect():2563] Redirects installed.
-2026-08-14 20:16:16,527 INFO    MainThread:409188 [wandb_init.py:init():1016] run started, returning control to user process
-2026-08-14 20:16:16,528 INFO    MainThread:409188 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 9, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'silu-waleed10', 'waleed_beta': 10.0, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-silu-waleed10-9L_run', 'per_device_train_batch_size': 512, 'num_train_epochs': 1, 'max_steps': 1000, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 1, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-silu-waleed10-9L-2.0M-20260814-201615', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 60000, 'eval_delay': 0, 'per_device_eval_batch_size': 512, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-glu-silu-waleed10-9L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
-2026-08-14 20:16:16,530 INFO    MainThread:409188 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 2001280 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x153f494c8950>>
-2026-08-14 20:16:16,530 INFO    MainThread:409188 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 2001280 None
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():820] starting backend
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():826] Connected to an existing wandb-core service via WANDB_SERVICE
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 20:21:28,415 INFO    MainThread:171901 [wandb_init.py:init():840] backend started and connected
+2026-08-14 20:21:28,416 INFO    MainThread:171901 [wandb_init.py:init():910] updated telemetry
+2026-08-14 20:21:28,423 INFO    MainThread:171901 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 20:21:28,685 INFO    MainThread:171901 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 20:21:28,763 INFO    MainThread:171901 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 20:21:28,763 INFO    MainThread:171901 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 20:21:28,763 INFO    MainThread:171901 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 20:21:28,763 INFO    MainThread:171901 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 20:21:28,764 INFO    MainThread:171901 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 20:21:28,765 INFO    MainThread:171901 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 100, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'waleed10', 'waleed_beta': 4, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-waleed10-100L_run', 'per_device_train_batch_size': 64, 'num_train_epochs': 1, 'max_steps': 2500, 'learning_rate': 0.0007, 'lr_scheduler_type': 'constant_with_warmup', 'lr_scheduler_kwargs': None, 'warmup_steps': 500, 'optim': 'adamw_torch_fused', 'optim_args': None, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 1, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-waleed10-100L-16.9M-20260814-202127', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 2498, 'eval_delay': 0, 'per_device_eval_batch_size': 1024, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/6L-glu-waleed10-100L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-14 20:21:28,771 INFO    MainThread:171901 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 16934016 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x1505b0edd050>>
+2026-08-14 20:21:28,771 INFO    MainThread:171901 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 16934016 None

zain/Activation/wandb/run-20260814_193907-kk1xbqht/logs/debug-core.log

@@ -20,3 +20,10 @@
 {"time":"2026-08-14T20:05:03.038763184Z","level":"INFO","msg":"handleInformInit: received","streamId":"bdhnr27j","id":"2(@)"}
 {"time":"2026-08-14T20:05:03.299131712Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"bdhnr27j","id":"2(@)"}
 {"time":"2026-08-14T20:05:08.644597551Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:26.598738734Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:27.243676424Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:27.245262581Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:21:27.245859065Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:21:28.155191217Z","level":"INFO","msg":"handleInformInit: received","streamId":"twgo1e89","id":"2(@)"}
+{"time":"2026-08-14T20:21:28.414618766Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"twgo1e89","id":"2(@)"}
+{"time":"2026-08-14T20:21:33.76572887Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"smlkpmpk76rp"}

zain/Activation/wandb/run-20260814_195151-ix6wcbwv/logs/debug-core.log

@@ -20,3 +20,10 @@
 {"time":"2026-08-14T20:05:03.038763184Z","level":"INFO","msg":"handleInformInit: received","streamId":"bdhnr27j","id":"2(@)"}
 {"time":"2026-08-14T20:05:03.299131712Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"bdhnr27j","id":"2(@)"}
 {"time":"2026-08-14T20:05:08.644597551Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:26.598738734Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:27.243676424Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:27.245262581Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:21:27.245859065Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:21:28.155191217Z","level":"INFO","msg":"handleInformInit: received","streamId":"twgo1e89","id":"2(@)"}
+{"time":"2026-08-14T20:21:28.414618766Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"twgo1e89","id":"2(@)"}
+{"time":"2026-08-14T20:21:33.76572887Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"smlkpmpk76rp"}

zain/Activation/wandb/run-20260814_200503-bdhnr27j/files/config.yaml

@@ -0,0 +1,437 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            k3jmuo8251i47dpekccsg2qseu6aglsk:
+                args:
+                    - --config
+                    - /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline100l.yaml
+                    - --variants
+                    - glu-relu
+                    - glu-gelu
+                    - glu-sigmoid
+                    - glu-waleed10
+                    - glu-silu-waleed10
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "696150671360"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 3fb47c6943d6927eafc39bd399a3d3a520bae74a
+                    remote: https://github.com/w-ahmad1a10/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation
+                startedAt: "2026-08-14T20:05:03.036402Z"
+                writerId: k3jmuo8251i47dpekccsg2qseu6aglsk
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 62
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.16.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: sigmoid
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 2498
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 1
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/6L-glu-sigmoid-100L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.0007
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant_with_warmup
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 2500
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: glu
+model/num_parameters:
+    value: 16934016
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 100
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch_fused
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: out/glu-sigmoid-100L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 1024
+per_device_train_batch_size:
+    value: 64
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: false
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-glu-sigmoid-100L-16.9M-20260814-200502
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.16.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+waleed_beta:
+    value: 4
+warmup_steps:
+    value: 500
+weight_decay:
+    value: 0

zain/Activation/wandb/run-20260814_200503-bdhnr27j/files/output.log

@@ -198,8 +198,64 @@ Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 14.
   - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
   - If you are not the owner of the model architecture class, please contact the model code owner to update it.
 Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 15.49it/s]
- 84%|████████▍ | 2097/2500 [12:54<02:37,  2.56it/s], ?it/s]
+ 84%|████████▍ | 2100/2500 [12:55<02:23,  2.78it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
 {'loss': '2.376', 'grad_norm': '0.4512', 'learning_rate': '0.0007', 'epoch': '0.1362', 'train/total_time_seconds': '639', 'train/time_per_step_avg': '0.3182', 'train/epoch_time_elapsed': '746.4', 'train/estimated_remaining_minutes': '2.531'}
 {'loss': '2.357', 'grad_norm': '0.4238', 'learning_rate': '0.0007', 'epoch': '0.1375', 'train/total_time_seconds': '645.4', 'train/time_per_step_avg': '0.3191', 'train/epoch_time_elapsed': '753.8', 'train/estimated_remaining_minutes': '2.426'}
 {'loss': '2.376', 'grad_norm': '0.4297', 'learning_rate': '0.0007', 'epoch': '0.1389', 'train/total_time_seconds': '651.7', 'train/time_per_step_avg': '0.319', 'train/epoch_time_elapsed': '761.2', 'train/estimated_remaining_minutes': '2.32'}
 {'loss': '2.349', 'grad_norm': '0.4336', 'learning_rate': '0.0007', 'epoch': '0.1402', 'train/total_time_seconds': '658.1', 'train/time_per_step_avg': '0.3182', 'train/epoch_time_elapsed': '768.6', 'train/estimated_remaining_minutes': '2.215'}
+{'loss': '2.34', 'grad_norm': '0.4395', 'learning_rate': '0.0007', 'epoch': '0.1416', 'train/total_time_seconds': '664.4', 'train/time_per_step_avg': '0.3186', 'train/epoch_time_elapsed': '775.9', 'train/estimated_remaining_minutes': '2.109'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 14.86it/s]
+ 88%|████████▊ | 2200/2500 [13:33<01:55,  2.60it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '2.346', 'grad_norm': '0.4297', 'learning_rate': '0.0007', 'epoch': '0.1429', 'train/total_time_seconds': '670.7', 'train/time_per_step_avg': '0.3173', 'train/epoch_time_elapsed': '783.7', 'train/estimated_remaining_minutes': '2.004'}
+{'loss': '2.35', 'grad_norm': '0.4492', 'learning_rate': '0.0007', 'epoch': '0.1443', 'train/total_time_seconds': '677.3', 'train/time_per_step_avg': '0.319', 'train/epoch_time_elapsed': '791.3', 'train/estimated_remaining_minutes': '1.899'}
+{'loss': '2.332', 'grad_norm': '0.4219', 'learning_rate': '0.0007', 'epoch': '0.1456', 'train/total_time_seconds': '683.7', 'train/time_per_step_avg': '0.3199', 'train/epoch_time_elapsed': '798.7', 'train/estimated_remaining_minutes': '1.794'}
+{'loss': '2.338', 'grad_norm': '0.4453', 'learning_rate': '0.0007', 'epoch': '0.1469', 'train/total_time_seconds': '690', 'train/time_per_step_avg': '0.3189', 'train/epoch_time_elapsed': '806', 'train/estimated_remaining_minutes': '1.688'}
+{'loss': '2.315', 'grad_norm': '0.4688', 'learning_rate': '0.0007', 'epoch': '0.1483', 'train/total_time_seconds': '696.5', 'train/time_per_step_avg': '0.3203', 'train/epoch_time_elapsed': '813.5', 'train/estimated_remaining_minutes': '1.583'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 14.25it/s]
+ 92%|█████████▏| 2300/2500 [14:10<01:11,  2.81it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '2.323', 'grad_norm': '0.4121', 'learning_rate': '0.0007', 'epoch': '0.1496', 'train/total_time_seconds': '702.7', 'train/time_per_step_avg': '0.3201', 'train/epoch_time_elapsed': '821', 'train/estimated_remaining_minutes': '1.477'}
+{'loss': '2.33', 'grad_norm': '0.4297', 'learning_rate': '0.0007', 'epoch': '0.151', 'train/total_time_seconds': '709.3', 'train/time_per_step_avg': '0.3199', 'train/epoch_time_elapsed': '828.6', 'train/estimated_remaining_minutes': '1.372'}
+{'loss': '2.308', 'grad_norm': '0.4082', 'learning_rate': '0.0007', 'epoch': '0.1523', 'train/total_time_seconds': '715.6', 'train/time_per_step_avg': '0.3191', 'train/epoch_time_elapsed': '835.9', 'train/estimated_remaining_minutes': '1.267'}
+{'loss': '2.319', 'grad_norm': '0.3945', 'learning_rate': '0.0007', 'epoch': '0.1537', 'train/total_time_seconds': '722.1', 'train/time_per_step_avg': '0.3205', 'train/epoch_time_elapsed': '843.4', 'train/estimated_remaining_minutes': '1.161'}
+{'loss': '2.298', 'grad_norm': '0.3887', 'learning_rate': '0.0007', 'epoch': '0.155', 'train/total_time_seconds': '728.4', 'train/time_per_step_avg': '0.3195', 'train/epoch_time_elapsed': '850.8', 'train/estimated_remaining_minutes': '1.056'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 12.03it/s]
+ 96%|█████████▌| 2400/2500 [14:48<00:37,  2.69it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '2.307', 'grad_norm': '0.4375', 'learning_rate': '0.0007', 'epoch': '0.1564', 'train/total_time_seconds': '734.8', 'train/time_per_step_avg': '0.3208', 'train/epoch_time_elapsed': '858.5', 'train/estimated_remaining_minutes': '0.9502'}
+{'loss': '2.3', 'grad_norm': '0.4512', 'learning_rate': '0.0007', 'epoch': '0.1577', 'train/total_time_seconds': '741.2', 'train/time_per_step_avg': '0.3188', 'train/epoch_time_elapsed': '865.9', 'train/estimated_remaining_minutes': '0.8447'}
+{'loss': '2.301', 'grad_norm': '0.4023', 'learning_rate': '0.0007', 'epoch': '0.1591', 'train/total_time_seconds': '747.5', 'train/time_per_step_avg': '0.3187', 'train/epoch_time_elapsed': '873.2', 'train/estimated_remaining_minutes': '0.739'}
+{'loss': '2.302', 'grad_norm': '0.4258', 'learning_rate': '0.0007', 'epoch': '0.1604', 'train/total_time_seconds': '753.8', 'train/time_per_step_avg': '0.3176', 'train/epoch_time_elapsed': '880.6', 'train/estimated_remaining_minutes': '0.6335'}
+{'loss': '2.282', 'grad_norm': '0.4453', 'learning_rate': '0.0007', 'epoch': '0.1618', 'train/total_time_seconds': '760.4', 'train/time_per_step_avg': '0.3194', 'train/epoch_time_elapsed': '888.1', 'train/estimated_remaining_minutes': '0.528'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 14.86it/s]
+100%|██████████| 2500/2500 [16:03<00:00,  4.37s[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '2.283', 'grad_norm': '0.4062', 'learning_rate': '0.0007', 'epoch': '0.1631', 'train/total_time_seconds': '766.7', 'train/time_per_step_avg': '0.3187', 'train/epoch_time_elapsed': '896', 'train/estimated_remaining_minutes': '0.4224'}
+{'loss': '2.273', 'grad_norm': '0.4102', 'learning_rate': '0.0007', 'epoch': '0.1645', 'train/total_time_seconds': '773.2', 'train/time_per_step_avg': '0.3204', 'train/epoch_time_elapsed': '903.5', 'train/estimated_remaining_minutes': '0.3169'}
+{'loss': '2.284', 'grad_norm': '0.4141', 'learning_rate': '0.0007', 'epoch': '0.1658', 'train/total_time_seconds': '779.5', 'train/time_per_step_avg': '0.3203', 'train/epoch_time_elapsed': '910.8', 'train/estimated_remaining_minutes': '0.2113'}
+{'loss': '2.282', 'grad_norm': '0.4375', 'learning_rate': '0.0007', 'epoch': '0.1672', 'train/total_time_seconds': '786', 'train/time_per_step_avg': '0.3215', 'train/epoch_time_elapsed': '918.4', 'train/estimated_remaining_minutes': '0.1056'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.28', 'eval_runtime': '19.09', 'eval_samples_per_second': '499', 'eval_steps_per_second': '0.524', 'epoch': '0.1684', 'train/total_time_seconds': '791.7', 'train/time_per_step_avg': '0.3193', 'train/epoch_time_elapsed': '944', 'train/estimated_remaining_minutes': '0.01056'}
+{'loss': '2.267', 'grad_norm': '0.4414', 'learning_rate': '0.0007', 'epoch': '0.1685', 'train/total_time_seconds': '792.3', 'train/time_per_step_avg': '0.3193', 'train/epoch_time_elapsed': '944.8', 'train/estimated_remaining_minutes': '0'}
+{'eval_loss': '2.279', 'eval_runtime': '19.09', 'eval_samples_per_second': '499.1', 'eval_steps_per_second': '0.524', 'epoch': '0.1685', 'train/total_time_seconds': '792.3', 'train/time_per_step_avg': '0.3193', 'train/epoch_time_elapsed': '963.9', 'train/estimated_remaining_minutes': '0'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 16.18it/s]
+100%|██████████| 2500/2500 [16:04<00:00,  2.59it/s], ?it/s]
+{'train_runtime': '964.9', 'train_samples_per_second': '165.8', 'train_steps_per_second': '2.591', 'train_loss': '3.169', 'epoch': '0.1685', 'train/total_time_seconds': '792.3', 'train/time_per_step_avg': '0.3193', 'train/epoch_time_elapsed': '964.2', 'train/estimated_remaining_minutes': '0'}
+100%|██████████| 10/10 [00:16<00:00,  1.63s/it]
+[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 16.29it/s]
+>>> FINISHED glu-sigmoid-100L successfully

zain/Activation/wandb/run-20260814_200503-bdhnr27j/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train/learning_rate":0.0007,"eval/loss":2.279484510421753,"train_samples_per_second":165.827,"train_loss":3.1693840744018553,"train/global_step":2500,"train/epoch":0.16852039096730703,"_wandb":{"runtime":983},"train_runtime":964.8638,"_timestamp":1.7867388865003955e+09,"_step":128,"eval/runtime":18.6737,"_runtime":983,"eval/steps_per_second":0.536,"total_flos":8.06570950656e+15,"train_steps_per_second":2.591,"train/grad_norm":0.44140625,"eval/samples_per_second":510.182,"train/loss":2.2665170669555663}

zain/Activation/wandb/run-20260814_200503-bdhnr27j/logs/debug-core.log

@@ -20,3 +20,10 @@
 {"time":"2026-08-14T20:05:03.038763184Z","level":"INFO","msg":"handleInformInit: received","streamId":"bdhnr27j","id":"2(@)"}
 {"time":"2026-08-14T20:05:03.299131712Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"bdhnr27j","id":"2(@)"}
 {"time":"2026-08-14T20:05:08.644597551Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:26.598738734Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:27.243676424Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:27.245262581Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:21:27.245859065Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:21:28.155191217Z","level":"INFO","msg":"handleInformInit: received","streamId":"twgo1e89","id":"2(@)"}
+{"time":"2026-08-14T20:21:28.414618766Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"twgo1e89","id":"2(@)"}
+{"time":"2026-08-14T20:21:33.76572887Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"smlkpmpk76rp"}

zain/Activation/wandb/run-20260814_200503-bdhnr27j/logs/debug-internal.log

@@ -109,3 +109,41 @@
 {"time":"2026-08-14T20:17:33.78015232Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-14T20:17:48.651274947Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":101,"history_lines":2,"events_offset":99,"events_lines":2,"console_offset":193,"console_lines":11}
 {"time":"2026-08-14T20:17:48.811733241Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:18:03.651316435Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":103,"history_lines":2,"events_offset":101,"events_lines":2,"console_offset":200,"console_lines":1}
+{"time":"2026-08-14T20:18:03.782036948Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:18:18.650637432Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":105,"history_lines":2,"events_offset":103,"events_lines":2,"console_offset":204,"console_lines":9}
+{"time":"2026-08-14T20:18:18.799509608Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:18:33.650976577Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":107,"history_lines":2,"events_offset":105,"events_lines":2,"console_offset":210,"console_lines":1}
+{"time":"2026-08-14T20:18:33.770841837Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:18:48.651118566Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":109,"history_lines":2,"events_offset":107,"events_lines":2,"console_offset":210,"console_lines":1}
+{"time":"2026-08-14T20:18:48.823981214Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:19:03.650533215Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":111,"history_lines":2,"events_offset":109,"events_lines":2,"console_offset":213,"console_lines":11}
+{"time":"2026-08-14T20:19:03.773526127Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:19:18.650959534Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":113,"history_lines":2,"events_offset":111,"events_lines":2,"console_offset":220,"console_lines":1}
+{"time":"2026-08-14T20:19:18.763313016Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:19:33.650486045Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":115,"history_lines":2,"events_offset":113,"events_lines":2,"console_offset":224,"console_lines":9}
+{"time":"2026-08-14T20:19:33.774648395Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:19:48.650787556Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":117,"history_lines":2,"events_offset":115,"events_lines":2,"console_offset":230,"console_lines":1}
+{"time":"2026-08-14T20:19:48.766885321Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:20:03.650573072Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":119,"history_lines":2,"events_offset":117,"events_lines":2,"console_offset":230,"console_lines":1}
+{"time":"2026-08-14T20:20:03.77511956Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:20:18.650437274Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":121,"history_lines":2,"events_offset":119,"events_lines":2,"console_offset":233,"console_lines":11}
+{"time":"2026-08-14T20:20:18.784751016Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:20:33.651147398Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":123,"history_lines":1,"events_offset":121,"events_lines":2,"console_offset":240,"console_lines":1}
+{"time":"2026-08-14T20:20:33.779513623Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:20:48.651196631Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":124,"history_lines":2,"events_offset":123,"events_lines":2,"console_offset":240,"console_lines":1}
+{"time":"2026-08-14T20:20:48.774142695Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:03.650659599Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":125,"events_lines":2,"console_offset":244,"console_lines":4}
+{"time":"2026-08-14T20:21:03.818182106Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:18.650764891Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":126,"history_lines":2,"events_offset":127,"events_lines":2,"console_offset":240,"console_lines":1}
+{"time":"2026-08-14T20:21:18.764248698Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:26.955877917Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-14T20:21:26.956118552Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":128,"history_lines":1,"events_offset":129,"events_lines":1,"console_offset":245,"console_lines":1,"uploaded_len":3}
+{"time":"2026-08-14T20:21:27.093848488Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:27.094552215Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":248,"console_lines":13,"complete":true,"exit_code":0}
+{"time":"2026-08-14T20:21:27.240839061Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:27.242082466Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-14T20:21:27.245305208Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-14T20:21:27.245331149Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-14T20:21:27.245427003Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-14T20:21:27.245431477Z","level":"INFO","msg":"stream: all finished"}

zain/Activation/wandb/run-20260814_200503-bdhnr27j/logs/debug.log

@@ -18,3 +18,8 @@ config: {'_wandb': {}}
 2026-08-14 20:05:03,643 INFO    MainThread:171901 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 100, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'sigmoid', 'waleed_beta': 4, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-sigmoid-100L_run', 'per_device_train_batch_size': 64, 'num_train_epochs': 1, 'max_steps': 2500, 'learning_rate': 0.0007, 'lr_scheduler_type': 'constant_with_warmup', 'lr_scheduler_kwargs': None, 'warmup_steps': 500, 'optim': 'adamw_torch_fused', 'optim_args': None, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 1, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-sigmoid-100L-16.9M-20260814-200502', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 2498, 'eval_delay': 0, 'per_device_eval_batch_size': 1024, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/6L-glu-sigmoid-100L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
 2026-08-14 20:05:03,649 INFO    MainThread:171901 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 16934016 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x1505b0258e90>>
 2026-08-14 20:05:03,649 INFO    MainThread:171901 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 16934016 None
+2026-08-14 20:21:26,597 INFO    MainThread:171901 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/research3/bdhnr27j
+2026-08-14 20:21:26,598 INFO    MainThread:171901 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-14 20:21:26,598 INFO    MainThread:171901 [wandb_run.py:_restore():2570] restore
+2026-08-14 20:21:26,598 INFO    MainThread:171901 [wandb_run.py:_restore():2576] restore done
+2026-08-14 20:21:27,244 INFO    MainThread:171901 [wandb_run.py:_footer_sync_info():3993] logging synced files

zain/Activation/wandb/run-20260814_200503-bdhnr27j/run-bdhnr27j.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa4265b9d4c7fc9b48937c829def591fe3bdd8527ea69354cdb1b540950c34cd
-size 720896
+oid sha256:06508c4efbed01d8187894f9f085f74e998fd8f828dd8fb2bcfa1e17c94933ef
+size 919856

zain/Activation/wandb/run-20260814_201615-w8mr4w6k/files/output.log

@@ -8,6 +8,48 @@
   - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
   - If you are not the owner of the model architecture class, please contact the model code owner to update it.
 Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 165.02it/s]
- 15%|█▌        | 151/1000 [01:44<09:09,  1.54it/s]?, ?it/s]
+ 20%|██        | 200/1000 [02:18<09:27,  1.41it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
 {'loss': '4.172', 'grad_norm': '0.3789', 'learning_rate': '0.001', 'epoch': '0.06469', 'train/total_time_seconds': '38.08', 'train/time_per_step_avg': '0.3169', 'train/epoch_time_elapsed': '83.18', 'train/estimated_remaining_minutes': '4.655'}
 {'loss': '3.941', 'grad_norm': '0.5508', 'learning_rate': '0.001', 'epoch': '0.07547', 'train/total_time_seconds': '44.47', 'train/time_per_step_avg': '0.3175', 'train/epoch_time_elapsed': '97.07', 'train/estimated_remaining_minutes': '4.553'}
+{'loss': '3.753', 'grad_norm': '0.8086', 'learning_rate': '0.001', 'epoch': '0.08625', 'train/total_time_seconds': '50.58', 'train/time_per_step_avg': '0.3154', 'train/epoch_time_elapsed': '110.5', 'train/estimated_remaining_minutes': '4.426'}
+{'loss': '3.603', 'grad_norm': '0.4766', 'learning_rate': '0.001', 'epoch': '0.09704', 'train/total_time_seconds': '57.04', 'train/time_per_step_avg': '0.3161', 'train/epoch_time_elapsed': '124.4', 'train/estimated_remaining_minutes': '4.331'}
+{'loss': '3.479', 'grad_norm': '0.5938', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '63.43', 'train/time_per_step_avg': '0.3168', 'train/epoch_time_elapsed': '138.3', 'train/estimated_remaining_minutes': '4.228'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 148.72it/s]
+ 30%|███       | 300/1000 [03:27<08:11,  1.43it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '3.364', 'grad_norm': '0.9297', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '69.8', 'train/time_per_step_avg': '0.3172', 'train/epoch_time_elapsed': '152.1', 'train/estimated_remaining_minutes': '4.125'}
+{'loss': '3.269', 'grad_norm': '0.7031', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '76.23', 'train/time_per_step_avg': '0.3175', 'train/epoch_time_elapsed': '166.1', 'train/estimated_remaining_minutes': '4.023'}
+{'loss': '3.183', 'grad_norm': '0.5039', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '82.4', 'train/time_per_step_avg': '0.3182', 'train/epoch_time_elapsed': '179.7', 'train/estimated_remaining_minutes': '3.909'}
+{'loss': '3.099', 'grad_norm': '0.6172', 'learning_rate': '0.001', 'epoch': '0.1509', 'train/total_time_seconds': '88.74', 'train/time_per_step_avg': '0.3169', 'train/epoch_time_elapsed': '193.5', 'train/estimated_remaining_minutes': '3.803'}
+{'loss': '3.024', 'grad_norm': '0.8359', 'learning_rate': '0.001', 'epoch': '0.1617', 'train/total_time_seconds': '95.06', 'train/time_per_step_avg': '0.3164', 'train/epoch_time_elapsed': '207.4', 'train/estimated_remaining_minutes': '3.697'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 153.87it/s]
+ 40%|████      | 400/1000 [04:35<06:49,  1.46it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '2.964', 'grad_norm': '0.5195', 'learning_rate': '0.001', 'epoch': '0.1725', 'train/total_time_seconds': '101.4', 'train/time_per_step_avg': '0.3156', 'train/epoch_time_elapsed': '221.3', 'train/estimated_remaining_minutes': '3.59'}
+{'loss': '2.899', 'grad_norm': '0.8398', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '107.9', 'train/time_per_step_avg': '0.3167', 'train/epoch_time_elapsed': '235.3', 'train/estimated_remaining_minutes': '3.491'}
+{'loss': '2.852', 'grad_norm': '0.4434', 'learning_rate': '0.001', 'epoch': '0.1941', 'train/total_time_seconds': '114.3', 'train/time_per_step_avg': '0.3187', 'train/epoch_time_elapsed': '249.2', 'train/estimated_remaining_minutes': '3.386'}
+{'loss': '2.79', 'grad_norm': '0.7227', 'learning_rate': '0.001', 'epoch': '0.2049', 'train/total_time_seconds': '119.8', 'train/time_per_step_avg': '0.3111', 'train/epoch_time_elapsed': '262.2', 'train/estimated_remaining_minutes': '3.259'}
+{'loss': '2.762', 'grad_norm': '0.8008', 'learning_rate': '0.001', 'epoch': '0.2156', 'train/total_time_seconds': '125.9', 'train/time_per_step_avg': '0.3088', 'train/epoch_time_elapsed': '275.8', 'train/estimated_remaining_minutes': '3.149'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 74.46it/s]
+ 50%|█████     | 500/1000 [05:43<05:36,  1.49it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '2.713', 'grad_norm': '0.832', 'learning_rate': '0.001', 'epoch': '0.2264', 'train/total_time_seconds': '132.2', 'train/time_per_step_avg': '0.3085', 'train/epoch_time_elapsed': '289.7', 'train/estimated_remaining_minutes': '3.043'}
+{'loss': '2.692', 'grad_norm': '0.668', 'learning_rate': '0.001', 'epoch': '0.2372', 'train/total_time_seconds': '137.5', 'train/time_per_step_avg': '0.296', 'train/epoch_time_elapsed': '302.5', 'train/estimated_remaining_minutes': '2.917'}
+{'loss': '2.655', 'grad_norm': '0.6992', 'learning_rate': '0.001', 'epoch': '0.248', 'train/total_time_seconds': '143.1', 'train/time_per_step_avg': '0.2887', 'train/epoch_time_elapsed': '315.7', 'train/estimated_remaining_minutes': '2.801'}
+{'loss': '2.621', 'grad_norm': '0.5898', 'learning_rate': '0.001', 'epoch': '0.2588', 'train/total_time_seconds': '149.5', 'train/time_per_step_avg': '0.2969', 'train/epoch_time_elapsed': '329.5', 'train/estimated_remaining_minutes': '2.7'}
+{'loss': '2.593', 'grad_norm': '0.8047', 'learning_rate': '0.001', 'epoch': '0.2695', 'train/total_time_seconds': '156', 'train/time_per_step_avg': '0.3', 'train/epoch_time_elapsed': '343.4', 'train/estimated_remaining_minutes': '2.599'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 169.74it/s]
+ 60%|█████▉    | 596/1000 [06:49<04:51,  1.39it/s]?, ?it/s]
+{'loss': '2.567', 'grad_norm': '0.8789', 'learning_rate': '0.001', 'epoch': '0.2803', 'train/total_time_seconds': '162.1', 'train/time_per_step_avg': '0.2986', 'train/epoch_time_elapsed': '356.9', 'train/estimated_remaining_minutes': '2.493'}
+{'loss': '2.552', 'grad_norm': '0.6875', 'learning_rate': '0.001', 'epoch': '0.2911', 'train/total_time_seconds': '168.4', 'train/time_per_step_avg': '0.3094', 'train/epoch_time_elapsed': '370.8', 'train/estimated_remaining_minutes': '2.391'}
+{'loss': '2.525', 'grad_norm': '0.5938', 'learning_rate': '0.001', 'epoch': '0.3019', 'train/total_time_seconds': '174.7', 'train/time_per_step_avg': '0.3159', 'train/epoch_time_elapsed': '384.6', 'train/estimated_remaining_minutes': '2.288'}
+{'loss': '2.504', 'grad_norm': '0.582', 'learning_rate': '0.001', 'epoch': '0.3127', 'train/total_time_seconds': '181.1', 'train/time_per_step_avg': '0.316', 'train/epoch_time_elapsed': '398.5', 'train/estimated_remaining_minutes': '2.186'}

zain/Activation/wandb/run-20260814_201615-w8mr4w6k/logs/debug-internal.log

@@ -21,3 +21,43 @@
 {"time":"2026-08-14T20:17:46.65675743Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-14T20:18:01.532320003Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":11,"events_lines":2,"console_offset":10,"console_lines":1}
 {"time":"2026-08-14T20:18:01.703907556Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:18:16.531644534Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":13,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:18:16.675676326Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:18:31.532290235Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":15,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:18:31.643700569Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:18:46.532044147Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":17,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:18:46.654285795Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:19:01.531775964Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":19,"events_lines":2,"console_offset":12,"console_lines":10}
+{"time":"2026-08-14T20:19:01.665198532Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:19:16.531677997Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":2,"events_offset":21,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:19:16.659350998Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:19:31.532173116Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":23,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:19:31.668108059Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:19:46.532116624Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":25,"events_lines":2,"console_offset":20,"console_lines":1}
+{"time":"2026-08-14T20:19:46.652602223Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:20:01.531504497Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":27,"events_lines":2,"console_offset":22,"console_lines":10}
+{"time":"2026-08-14T20:20:01.662861652Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:20:16.531672459Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":29,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T20:20:16.710856839Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:20:31.531798457Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":31,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T20:20:31.678937064Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:20:46.53201641Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":33,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T20:20:46.674672615Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:01.532133559Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":35,"events_lines":2,"console_offset":30,"console_lines":1}
+{"time":"2026-08-14T20:21:01.65388367Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:16.531525652Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":37,"events_lines":2,"console_offset":32,"console_lines":10}
+{"time":"2026-08-14T20:21:16.653161683Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:31.532412778Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":39,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T20:21:31.667507491Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:46.532379533Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":2,"events_offset":41,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T20:21:46.644719966Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:01.531517432Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":43,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-14T20:22:01.726312705Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:16.531867433Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":45,"events_lines":2,"console_offset":42,"console_lines":10}
+{"time":"2026-08-14T20:22:16.759084254Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:31.532145449Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":47,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T20:22:31.664579953Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:46.53157997Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":49,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T20:22:46.664689895Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:23:01.532613131Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":51,"events_lines":2,"console_offset":50,"console_lines":1}
+{"time":"2026-08-14T20:23:01.736637578Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260814_202128-twgo1e89/files/output.log

@@ -0,0 +1,24 @@
+  4%|▍         | 100/2500 [00:37<14:24,  2.78it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '8.299', 'grad_norm': '2.125', 'learning_rate': '2.66e-05', 'epoch': '0.001348', 'train/total_time_seconds': '6.766', 'train/time_per_step_avg': '0.3383', 'train/epoch_time_elapsed': '7.843', 'train/estimated_remaining_minutes': '13.98'}
+{'loss': '8.036', 'grad_norm': '1.305', 'learning_rate': '5.46e-05', 'epoch': '0.002696', 'train/total_time_seconds': '13.31', 'train/time_per_step_avg': '0.3328', 'train/epoch_time_elapsed': '15.4', 'train/estimated_remaining_minutes': '13.64'}
+{'loss': '7.835', 'grad_norm': '1.273', 'learning_rate': '8.26e-05', 'epoch': '0.004044', 'train/total_time_seconds': '19.82', 'train/time_per_step_avg': '0.3303', 'train/epoch_time_elapsed': '22.9', 'train/estimated_remaining_minutes': '13.43'}
+{'loss': '7.572', 'grad_norm': '1.266', 'learning_rate': '0.0001106', 'epoch': '0.005393', 'train/total_time_seconds': '26.21', 'train/time_per_step_avg': '0.3276', 'train/epoch_time_elapsed': '30.3', 'train/estimated_remaining_minutes': '13.21'}
+{'loss': '7.237', 'grad_norm': '1.195', 'learning_rate': '0.0001386', 'epoch': '0.006741', 'train/total_time_seconds': '32.57', 'train/time_per_step_avg': '0.3257', 'train/epoch_time_elapsed': '37.66', 'train/estimated_remaining_minutes': '13.03'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 16.15it/s]
+  8%|▊         | 200/2500 [01:15<14:25,  2.66it/s][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '6.865', 'grad_norm': '1.148', 'learning_rate': '0.0001666', 'epoch': '0.008089', 'train/total_time_seconds': '39.05', 'train/time_per_step_avg': '0.3228', 'train/epoch_time_elapsed': '45.47', 'train/estimated_remaining_minutes': '12.91'}
+{'loss': '6.51', 'grad_norm': '1.023', 'learning_rate': '0.0001946', 'epoch': '0.009437', 'train/total_time_seconds': '45.44', 'train/time_per_step_avg': '0.3213', 'train/epoch_time_elapsed': '52.86', 'train/estimated_remaining_minutes': '12.77'}
+{'loss': '6.204', 'grad_norm': '1.031', 'learning_rate': '0.0002226', 'epoch': '0.01079', 'train/total_time_seconds': '51.9', 'train/time_per_step_avg': '0.3208', 'train/epoch_time_elapsed': '60.39', 'train/estimated_remaining_minutes': '12.65'}
+{'loss': '5.941', 'grad_norm': '0.8711', 'learning_rate': '0.0002506', 'epoch': '0.01213', 'train/total_time_seconds': '58.3', 'train/time_per_step_avg': '0.3209', 'train/epoch_time_elapsed': '67.79', 'train/estimated_remaining_minutes': '12.52'}
+{'loss': '5.677', 'grad_norm': '0.7617', 'learning_rate': '0.0002786', 'epoch': '0.01348', 'train/total_time_seconds': '64.86', 'train/time_per_step_avg': '0.3229', 'train/epoch_time_elapsed': '75.34', 'train/estimated_remaining_minutes': '12.43'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 17.17it/s]
+ 11%|█         | 264/2500 [01:39<13:33,  2.75it/s]?, ?it/s]
+{'loss': '5.419', 'grad_norm': '0.6836', 'learning_rate': '0.0003066', 'epoch': '0.01483', 'train/total_time_seconds': '71.42', 'train/time_per_step_avg': '0.3238', 'train/epoch_time_elapsed': '83.21', 'train/estimated_remaining_minutes': '12.34'}
+{'loss': '5.139', 'grad_norm': '0.6562', 'learning_rate': '0.0003346', 'epoch': '0.01618', 'train/total_time_seconds': '77.85', 'train/time_per_step_avg': '0.3241', 'train/epoch_time_elapsed': '90.64', 'train/estimated_remaining_minutes': '12.22'}
+{'loss': '4.887', 'grad_norm': '1.266', 'learning_rate': '0.0003626', 'epoch': '0.01753', 'train/total_time_seconds': '84.47', 'train/time_per_step_avg': '0.3258', 'train/epoch_time_elapsed': '98.27', 'train/estimated_remaining_minutes': '12.13'}

zain/Activation/wandb/run-20260814_202128-twgo1e89/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.16.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

zain/Activation/wandb/run-20260814_202128-twgo1e89/files/wandb-metadata.json

@@ -0,0 +1,99 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-14T20:21:28.152860Z",
+  "args":  [
+    "--config",
+    "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/configs/baseline100l.yaml",
+    "--variants",
+    "glu-relu",
+    "glu-gelu",
+    "glu-sigmoid",
+    "glu-waleed10",
+    "glu-silu-waleed10"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/w-ahmad1a10/Activation.git",
+    "commit":  "3fb47c6943d6927eafc39bd399a3d3a520bae74a"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "696168878080"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "d8dewmrsjbytqiua7y7ndqbzi7lmdrya"
+}

zain/Activation/wandb/run-20260814_202128-twgo1e89/logs/debug-core.log

@@ -0,0 +1,29 @@
+{"time":"2026-08-14T19:38:51.636058907Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmp54729boe/port-1375539.txt","pid":1375539,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-14T19:38:51.639970993Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":1375539}
+{"time":"2026-08-14T19:38:51.639940209Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-1375539-170701-251437063/socket","Net":"unix"}}
+{"time":"2026-08-14T19:38:51.812900736Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-14T19:39:07.276688434Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"2(@)"}
+{"time":"2026-08-14T19:39:07.370077387Z","level":"INFO","msg":"handleInformInit: received","streamId":"kk1xbqht","id":"2(@)"}
+{"time":"2026-08-14T19:39:07.640206016Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"kk1xbqht","id":"2(@)"}
+{"time":"2026-08-14T19:39:13.074249733Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"zj3lg1k64oen"}
+{"time":"2026-08-14T19:51:49.732364673Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"zj3lg1k64oen"}
+{"time":"2026-08-14T19:51:50.223490459Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"zj3lg1k64oen"}
+{"time":"2026-08-14T19:51:50.225275352Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"kk1xbqht","id":"2(@)"}
+{"time":"2026-08-14T19:51:50.225791347Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"kk1xbqht","id":"2(@)"}
+{"time":"2026-08-14T19:51:51.140595835Z","level":"INFO","msg":"handleInformInit: received","streamId":"ix6wcbwv","id":"2(@)"}
+{"time":"2026-08-14T19:51:51.407779103Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"ix6wcbwv","id":"2(@)"}
+{"time":"2026-08-14T19:51:56.793135206Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"m2u1nqrk4r4j"}
+{"time":"2026-08-14T20:05:01.389026301Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"m2u1nqrk4r4j"}
+{"time":"2026-08-14T20:05:01.891785946Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"m2u1nqrk4r4j"}
+{"time":"2026-08-14T20:05:01.893356932Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"ix6wcbwv","id":"2(@)"}
+{"time":"2026-08-14T20:05:01.894067235Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"ix6wcbwv","id":"2(@)"}
+{"time":"2026-08-14T20:05:03.038763184Z","level":"INFO","msg":"handleInformInit: received","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:05:03.299131712Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:05:08.644597551Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:26.598738734Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:27.243676424Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"u5vxhmn1tauv"}
+{"time":"2026-08-14T20:21:27.245262581Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:21:27.245859065Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"bdhnr27j","id":"2(@)"}
+{"time":"2026-08-14T20:21:28.155191217Z","level":"INFO","msg":"handleInformInit: received","streamId":"twgo1e89","id":"2(@)"}
+{"time":"2026-08-14T20:21:28.414618766Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"twgo1e89","id":"2(@)"}
+{"time":"2026-08-14T20:21:33.76572887Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"smlkpmpk76rp"}

zain/Activation/wandb/run-20260814_202128-twgo1e89/logs/debug-internal.log

@@ -0,0 +1,21 @@
+{"time":"2026-08-14T20:21:28.155354055Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-14T20:21:28.155495403Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-14T20:21:28.41445694Z","level":"INFO","msg":"stream: created new stream","id":"twgo1e89"}
+{"time":"2026-08-14T20:21:28.41452958Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-14T20:21:28.414611994Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-14T20:21:28.414621977Z","level":"INFO","msg":"writer: started","stream_id":"twgo1e89"}
+{"time":"2026-08-14T20:21:28.414641036Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-14T20:21:28.77231947Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:21:28.885103894Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:43.772746583Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":0,"events_lines":1,"console_offset":0,"console_lines":2,"uploaded_len":2}
+{"time":"2026-08-14T20:21:43.910826986Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:21:58.773175035Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":2,"events_offset":1,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:21:58.957050518Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:13.773475474Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":2,"events_offset":3,"events_lines":2,"console_offset":0,"console_lines":1}
+{"time":"2026-08-14T20:22:13.977519477Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:28.773095938Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":2,"events_offset":5,"events_lines":2,"console_offset":2,"console_lines":11}
+{"time":"2026-08-14T20:22:28.955788633Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:43.773237697Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":2,"events_offset":7,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:22:43.926628755Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-14T20:22:58.772568445Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":2,"events_offset":9,"events_lines":2,"console_offset":10,"console_lines":1}
+{"time":"2026-08-14T20:22:58.931284477Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260814_202128-twgo1e89/logs/debug.log

@@ -0,0 +1,20 @@
+2026-08-14 20:21:28,153 INFO    MainThread:171901 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_202128-twgo1e89/logs/debug.log
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/zain/Activation/wandb/run-20260814_202128-twgo1e89/logs/debug-internal.log
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():772] calling init triggers
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():820] starting backend
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():826] Connected to an existing wandb-core service via WANDB_SERVICE
+2026-08-14 20:21:28,154 INFO    MainThread:171901 [wandb_init.py:init():835] sending inform_init request
+2026-08-14 20:21:28,415 INFO    MainThread:171901 [wandb_init.py:init():840] backend started and connected
+2026-08-14 20:21:28,416 INFO    MainThread:171901 [wandb_init.py:init():910] updated telemetry
+2026-08-14 20:21:28,423 INFO    MainThread:171901 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-14 20:21:28,685 INFO    MainThread:171901 [wandb_init.py:init():978] starting run threads in backend
+2026-08-14 20:21:28,763 INFO    MainThread:171901 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-14 20:21:28,763 INFO    MainThread:171901 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-14 20:21:28,763 INFO    MainThread:171901 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-14 20:21:28,763 INFO    MainThread:171901 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-14 20:21:28,764 INFO    MainThread:171901 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-14 20:21:28,765 INFO    MainThread:171901 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.16.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 100, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'waleed10', 'waleed_beta': 4, 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-waleed10-100L_run', 'per_device_train_batch_size': 64, 'num_train_epochs': 1, 'max_steps': 2500, 'learning_rate': 0.0007, 'lr_scheduler_type': 'constant_with_warmup', 'lr_scheduler_kwargs': None, 'warmup_steps': 500, 'optim': 'adamw_torch_fused', 'optim_args': None, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 1, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-waleed10-100L-16.9M-20260814-202127', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 2498, 'eval_delay': 0, 'per_device_eval_batch_size': 1024, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': False, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/6L-glu-waleed10-100L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-14 20:21:28,771 INFO    MainThread:171901 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 16934016 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x1505b0edd050>>
+2026-08-14 20:21:28,771 INFO    MainThread:171901 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 16934016 None