Training in progress, epoch 4

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ddac4d3a3d0da455d0e501fd182e68910ab7b0a78868896b9607f09a4a31cd3
+oid sha256:37d04ba8e33af7df58f2833c2680355fd52b2929329d03cbf34ac886408204f4
 size 437958648

run-0/checkpoint-117/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8672e581da9d8df927955f18ee97b1fb7bc4b72d39bba7c191508198135749ab
+oid sha256:f31da7ed19c4d54724eff5814fc70fc1b0359968b8eb5455d2363940d09056e5
 size 437958648

run-0/checkpoint-117/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7c35b295bba54ba1b241603995ce1811c2382076f99713945e46fceacbb9937
+oid sha256:727bb5dc729ab51a1b4c6c405bb4f4ea1a84b00ef487b3407e16a1530da98d24
 size 876038394

run-0/checkpoint-117/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9df7fcee919248151322838e3ee2bda70121eb4c651bc0be349c4f5f62deeaf
+oid sha256:8dedec8cb35bb4edec4b57e62ce9549f5bcd5baa05a7b09482e7f936fba95bd7
 size 5432

run-0/checkpoint-156/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1543d11d39ae4983ad4b56210a391fd44e53d905d92d2536f8ce4494e3795db5
+oid sha256:37d04ba8e33af7df58f2833c2680355fd52b2929329d03cbf34ac886408204f4
 size 437958648

run-0/checkpoint-156/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:889253018f7f4ef7fe5b3708a9b99fd213e993b42054f01aab66786ee4ea3395
+oid sha256:0dbe2ee4a6c940354ab98038be65bd5e28a470d52963a8ac58d946eea229ef37
 size 876038394

run-0/checkpoint-156/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 156,
-  "best_metric": 0.6967509025270758,
+  "best_metric": 0.7184115523465704,
   "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-156",
   "epoch": 4.0,
   "eval_steps": 500,
@@ -11,143 +11,143 @@
   "log_history": [
     {
       "epoch": 0.2564102564102564,
-      "grad_norm": 1.662626028060913,
+      "grad_norm": 1.662625789642334,
       "learning_rate": 9.487179487179487e-05,
       "loss": 0.696,
       "step": 10
     },
     {
       "epoch": 0.5128205128205128,
-      "grad_norm": 2.0300467014312744,
+      "grad_norm": 2.0300142765045166,
       "learning_rate": 8.974358974358975e-05,
       "loss": 0.6793,
       "step": 20
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 4.490738868713379,
+      "grad_norm": 4.4956440925598145,
       "learning_rate": 8.461538461538461e-05,
       "loss": 0.6499,
       "step": 30
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.631768953068592,
-      "eval_loss": 0.6309793591499329,
-      "eval_runtime": 0.6606,
-      "eval_samples_per_second": 419.298,
-      "eval_steps_per_second": 7.569,
+      "eval_accuracy": 0.6173285198555957,
+      "eval_loss": 0.6330550909042358,
+      "eval_runtime": 0.6611,
+      "eval_samples_per_second": 418.98,
+      "eval_steps_per_second": 7.563,
       "step": 39
     },
     {
       "epoch": 1.0256410256410255,
-      "grad_norm": 3.481339454650879,
+      "grad_norm": 3.44403338432312,
       "learning_rate": 7.948717948717948e-05,
-      "loss": 0.6793,
+      "loss": 0.6782,
       "step": 40
     },
     {
       "epoch": 1.282051282051282,
-      "grad_norm": 4.932971477508545,
+      "grad_norm": 5.425619602203369,
       "learning_rate": 7.435897435897436e-05,
-      "loss": 0.5602,
+      "loss": 0.5831,
       "step": 50
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 7.099682807922363,
+      "grad_norm": 11.772316932678223,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.5998,
+      "loss": 0.5307,
       "step": 60
     },
     {
       "epoch": 1.7948717948717947,
-      "grad_norm": 3.696152925491333,
+      "grad_norm": 3.624584674835205,
       "learning_rate": 6.410256410256412e-05,
-      "loss": 0.5403,
+      "loss": 0.5396,
       "step": 70
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6931407942238267,
-      "eval_loss": 0.6287456154823303,
-      "eval_runtime": 0.6619,
-      "eval_samples_per_second": 418.511,
-      "eval_steps_per_second": 7.554,
+      "eval_accuracy": 0.6859205776173285,
+      "eval_loss": 0.6415449380874634,
+      "eval_runtime": 0.6613,
+      "eval_samples_per_second": 418.848,
+      "eval_steps_per_second": 7.56,
       "step": 78
     },
     {
       "epoch": 2.051282051282051,
-      "grad_norm": 3.8591785430908203,
+      "grad_norm": 3.940481662750244,
       "learning_rate": 5.897435897435898e-05,
-      "loss": 0.4872,
+      "loss": 0.455,
       "step": 80
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 4.27885627746582,
+      "grad_norm": 5.196528434753418,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.3396,
+      "loss": 0.3532,
       "step": 90
     },
     {
       "epoch": 2.564102564102564,
-      "grad_norm": 4.205892562866211,
+      "grad_norm": 4.290834903717041,
       "learning_rate": 4.871794871794872e-05,
-      "loss": 0.3372,
+      "loss": 0.3482,
       "step": 100
     },
     {
       "epoch": 2.8205128205128203,
-      "grad_norm": 5.800762176513672,
+      "grad_norm": 4.902711391448975,
       "learning_rate": 4.358974358974359e-05,
-      "loss": 0.3422,
+      "loss": 0.3223,
       "step": 110
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6787003610108303,
-      "eval_loss": 0.7227063179016113,
-      "eval_runtime": 0.6696,
-      "eval_samples_per_second": 413.706,
-      "eval_steps_per_second": 7.468,
+      "eval_accuracy": 0.6931407942238267,
+      "eval_loss": 0.7661525011062622,
+      "eval_runtime": 0.6709,
+      "eval_samples_per_second": 412.863,
+      "eval_steps_per_second": 7.452,
       "step": 117
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 3.5253195762634277,
+      "grad_norm": 4.413120269775391,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.2668,
+      "loss": 0.2385,
       "step": 120
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 11.2240629196167,
+      "grad_norm": 4.013213157653809,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.1835,
+      "loss": 0.1954,
       "step": 130
     },
     {
       "epoch": 3.58974358974359,
-      "grad_norm": 5.6398420333862305,
+      "grad_norm": 5.019925117492676,
       "learning_rate": 2.8205128205128207e-05,
-      "loss": 0.1725,
+      "loss": 0.1789,
       "step": 140
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 6.835488319396973,
+      "grad_norm": 4.320362091064453,
       "learning_rate": 2.307692307692308e-05,
-      "loss": 0.1297,
+      "loss": 0.1594,
       "step": 150
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6967509025270758,
-      "eval_loss": 0.9852063655853271,
-      "eval_runtime": 0.6608,
-      "eval_samples_per_second": 419.212,
-      "eval_steps_per_second": 7.567,
+      "eval_accuracy": 0.7184115523465704,
+      "eval_loss": 0.8570895791053772,
+      "eval_runtime": 0.6607,
+      "eval_samples_per_second": 419.255,
+      "eval_steps_per_second": 7.568,
       "step": 156
     }
   ],

run-0/checkpoint-156/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9df7fcee919248151322838e3ee2bda70121eb4c651bc0be349c4f5f62deeaf
+oid sha256:8dedec8cb35bb4edec4b57e62ce9549f5bcd5baa05a7b09482e7f936fba95bd7
 size 5432

run-0/checkpoint-195/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 156,
-  "best_metric": 0.6967509025270758,
+  "best_metric": 0.7184115523465704,
   "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-156",
   "epoch": 5.0,
   "eval_steps": 500,
@@ -11,180 +11,180 @@
   "log_history": [
     {
       "epoch": 0.2564102564102564,
-      "grad_norm": 1.662626028060913,
+      "grad_norm": 1.662625789642334,
       "learning_rate": 9.487179487179487e-05,
       "loss": 0.696,
       "step": 10
     },
     {
       "epoch": 0.5128205128205128,
-      "grad_norm": 2.0300467014312744,
+      "grad_norm": 2.0300142765045166,
       "learning_rate": 8.974358974358975e-05,
       "loss": 0.6793,
       "step": 20
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 4.490738868713379,
+      "grad_norm": 4.4956440925598145,
       "learning_rate": 8.461538461538461e-05,
       "loss": 0.6499,
       "step": 30
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.631768953068592,
-      "eval_loss": 0.6309793591499329,
-      "eval_runtime": 0.6606,
-      "eval_samples_per_second": 419.298,
-      "eval_steps_per_second": 7.569,
+      "eval_accuracy": 0.6173285198555957,
+      "eval_loss": 0.6330550909042358,
+      "eval_runtime": 0.6611,
+      "eval_samples_per_second": 418.98,
+      "eval_steps_per_second": 7.563,
       "step": 39
     },
     {
       "epoch": 1.0256410256410255,
-      "grad_norm": 3.481339454650879,
+      "grad_norm": 3.44403338432312,
       "learning_rate": 7.948717948717948e-05,
-      "loss": 0.6793,
+      "loss": 0.6782,
       "step": 40
     },
     {
       "epoch": 1.282051282051282,
-      "grad_norm": 4.932971477508545,
+      "grad_norm": 5.425619602203369,
       "learning_rate": 7.435897435897436e-05,
-      "loss": 0.5602,
+      "loss": 0.5831,
       "step": 50
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 7.099682807922363,
+      "grad_norm": 11.772316932678223,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.5998,
+      "loss": 0.5307,
       "step": 60
     },
     {
       "epoch": 1.7948717948717947,
-      "grad_norm": 3.696152925491333,
+      "grad_norm": 3.624584674835205,
       "learning_rate": 6.410256410256412e-05,
-      "loss": 0.5403,
+      "loss": 0.5396,
       "step": 70
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6931407942238267,
-      "eval_loss": 0.6287456154823303,
-      "eval_runtime": 0.6619,
-      "eval_samples_per_second": 418.511,
-      "eval_steps_per_second": 7.554,
+      "eval_accuracy": 0.6859205776173285,
+      "eval_loss": 0.6415449380874634,
+      "eval_runtime": 0.6613,
+      "eval_samples_per_second": 418.848,
+      "eval_steps_per_second": 7.56,
       "step": 78
     },
     {
       "epoch": 2.051282051282051,
-      "grad_norm": 3.8591785430908203,
+      "grad_norm": 3.940481662750244,
       "learning_rate": 5.897435897435898e-05,
-      "loss": 0.4872,
+      "loss": 0.455,
       "step": 80
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 4.27885627746582,
+      "grad_norm": 5.196528434753418,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.3396,
+      "loss": 0.3532,
       "step": 90
     },
     {
       "epoch": 2.564102564102564,
-      "grad_norm": 4.205892562866211,
+      "grad_norm": 4.290834903717041,
       "learning_rate": 4.871794871794872e-05,
-      "loss": 0.3372,
+      "loss": 0.3482,
       "step": 100
     },
     {
       "epoch": 2.8205128205128203,
-      "grad_norm": 5.800762176513672,
+      "grad_norm": 4.902711391448975,
       "learning_rate": 4.358974358974359e-05,
-      "loss": 0.3422,
+      "loss": 0.3223,
       "step": 110
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6787003610108303,
-      "eval_loss": 0.7227063179016113,
-      "eval_runtime": 0.6696,
-      "eval_samples_per_second": 413.706,
-      "eval_steps_per_second": 7.468,
+      "eval_accuracy": 0.6931407942238267,
+      "eval_loss": 0.7661525011062622,
+      "eval_runtime": 0.6709,
+      "eval_samples_per_second": 412.863,
+      "eval_steps_per_second": 7.452,
       "step": 117
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 3.5253195762634277,
+      "grad_norm": 4.413120269775391,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.2668,
+      "loss": 0.2385,
       "step": 120
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 11.2240629196167,
+      "grad_norm": 4.013213157653809,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.1835,
+      "loss": 0.1954,
       "step": 130
     },
     {
       "epoch": 3.58974358974359,
-      "grad_norm": 5.6398420333862305,
+      "grad_norm": 5.019925117492676,
       "learning_rate": 2.8205128205128207e-05,
-      "loss": 0.1725,
+      "loss": 0.1789,
       "step": 140
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 6.835488319396973,
+      "grad_norm": 4.320362091064453,
       "learning_rate": 2.307692307692308e-05,
-      "loss": 0.1297,
+      "loss": 0.1594,
       "step": 150
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6967509025270758,
-      "eval_loss": 0.9852063655853271,
-      "eval_runtime": 0.6608,
-      "eval_samples_per_second": 419.212,
-      "eval_steps_per_second": 7.567,
+      "eval_accuracy": 0.7184115523465704,
+      "eval_loss": 0.8570895791053772,
+      "eval_runtime": 0.6607,
+      "eval_samples_per_second": 419.255,
+      "eval_steps_per_second": 7.568,
       "step": 156
     },
     {
       "epoch": 4.102564102564102,
-      "grad_norm": 1.7055132389068604,
+      "grad_norm": 3.1741435527801514,
       "learning_rate": 1.794871794871795e-05,
-      "loss": 0.1088,
+      "loss": 0.126,
       "step": 160
     },
     {
       "epoch": 4.358974358974359,
-      "grad_norm": 4.614296913146973,
+      "grad_norm": 3.261113405227661,
       "learning_rate": 1.282051282051282e-05,
-      "loss": 0.0803,
+      "loss": 0.0947,
       "step": 170
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 4.053183555603027,
+      "grad_norm": 4.431177616119385,
       "learning_rate": 7.692307692307694e-06,
-      "loss": 0.0625,
+      "loss": 0.0828,
       "step": 180
     },
     {
       "epoch": 4.871794871794872,
-      "grad_norm": 7.156663417816162,
+      "grad_norm": 3.911816120147705,
       "learning_rate": 2.564102564102564e-06,
-      "loss": 0.0686,
+      "loss": 0.0922,
       "step": 190
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6859205776173285,
-      "eval_loss": 1.0840200185775757,
-      "eval_runtime": 0.6626,
-      "eval_samples_per_second": 418.066,
-      "eval_steps_per_second": 7.546,
+      "eval_accuracy": 0.7148014440433214,
+      "eval_loss": 1.0289418697357178,
+      "eval_runtime": 0.6615,
+      "eval_samples_per_second": 418.719,
+      "eval_steps_per_second": 7.558,
       "step": 195
     }
   ],

run-0/checkpoint-78/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24b4648ded986884eb769e5736bda3f39ff6235308b7ff65c5d9cae456d76b11
+oid sha256:acac08e7f9040f2f5af0d8e30a90a080ec7d11a0e00891bf50443f9385f53d7e
 size 437958648

run-0/checkpoint-78/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a6bd0bd4447c86fa907715bb20e5259062d2eb35ad8c4c9ae594e0c25acd0c2
+oid sha256:874e6ef84296305e210837a0dc28b56f9a3287e03ba609c676f01babdf78c2a8
 size 876038394