Upload folder using huggingface_hub

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9249496613582281
+      value: 0.918542925132711
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bert-large-cased](https://huggingface.co/bert-large-cased) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3051
-- Accuracy: 0.9249
+- Loss: 0.4854
+- Accuracy: 0.9185
 
 ## Model description
 
@@ -53,12 +53,12 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 32
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3.0
+- num_epochs: 5.0
 
 ### Training results
 

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.9249496613582281,
-    "eval_loss": 0.30514439940452576,
-    "eval_runtime": 31.6611,
+    "epoch": 5.0,
+    "eval_accuracy": 0.918542925132711,
+    "eval_loss": 0.4854465425014496,
+    "eval_runtime": 13.9725,
     "eval_samples": 5463,
-    "eval_samples_per_second": 172.546,
-    "eval_steps_per_second": 21.572,
-    "train_loss": 0.18336697417057063,
-    "train_runtime": 5113.4763,
+    "eval_samples_per_second": 390.982,
+    "eval_steps_per_second": 48.882,
+    "train_loss": 0.16656628906298943,
+    "train_runtime": 2957.7214,
     "train_samples": 104743,
-    "train_samples_per_second": 61.451,
-    "train_steps_per_second": 1.921
+    "train_samples_per_second": 177.067,
+    "train_steps_per_second": 11.068
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.9249496613582281,
-    "eval_loss": 0.30514439940452576,
-    "eval_runtime": 31.6611,
+    "epoch": 5.0,
+    "eval_accuracy": 0.918542925132711,
+    "eval_loss": 0.4854465425014496,
+    "eval_runtime": 13.9725,
     "eval_samples": 5463,
-    "eval_samples_per_second": 172.546,
-    "eval_steps_per_second": 21.572
+    "eval_samples_per_second": 390.982,
+    "eval_steps_per_second": 48.882
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1265b425013acfd717cea77a21097580cac4f7c9f5ac7e933746dc3c1d28b661
+oid sha256:cca9964db61787a087febc17a4027ae31c8cc5e76f622cde4d05465e2454591d
 size 1334459697

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "train_loss": 0.18336697417057063,
-    "train_runtime": 5113.4763,
+    "epoch": 5.0,
+    "train_loss": 0.16656628906298943,
+    "train_runtime": 2957.7214,
     "train_samples": 104743,
-    "train_samples_per_second": 61.451,
-    "train_steps_per_second": 1.921
+    "train_samples_per_second": 177.067,
+    "train_steps_per_second": 11.068
 }

trainer_state.json

@@ -1,139 +1,415 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.0,
-  "global_step": 9822,
+  "epoch": 5.0,
+  "global_step": 32735,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.15,
-      "learning_rate": 1.8981877418041133e-05,
-      "loss": 0.4196,
+      "epoch": 0.08,
+      "learning_rate": 1.9697571406751184e-05,
+      "loss": 0.4892,
       "step": 500
     },
     {
-      "epoch": 0.31,
-      "learning_rate": 1.7963754836082265e-05,
-      "loss": 0.3233,
+      "epoch": 0.15,
+      "learning_rate": 1.9392087979227128e-05,
+      "loss": 0.4021,
       "step": 1000
     },
     {
-      "epoch": 0.46,
-      "learning_rate": 1.69456322541234e-05,
-      "loss": 0.2968,
+      "epoch": 0.23,
+      "learning_rate": 1.908660455170307e-05,
+      "loss": 0.3816,
       "step": 1500
     },
     {
-      "epoch": 0.61,
-      "learning_rate": 1.592750967216453e-05,
-      "loss": 0.2858,
+      "epoch": 0.31,
+      "learning_rate": 1.8781121124179017e-05,
+      "loss": 0.3486,
       "step": 2000
     },
     {
-      "epoch": 0.76,
-      "learning_rate": 1.4909387090205662e-05,
-      "loss": 0.2731,
+      "epoch": 0.38,
+      "learning_rate": 1.8475637696654957e-05,
+      "loss": 0.335,
       "step": 2500
     },
     {
-      "epoch": 0.92,
-      "learning_rate": 1.3891264508246794e-05,
-      "loss": 0.257,
+      "epoch": 0.46,
+      "learning_rate": 1.81701542691309e-05,
+      "loss": 0.3268,
       "step": 3000
     },
     {
-      "epoch": 1.07,
-      "learning_rate": 1.2873141926287927e-05,
-      "loss": 0.2154,
+      "epoch": 0.53,
+      "learning_rate": 1.786528180846189e-05,
+      "loss": 0.3362,
       "step": 3500
     },
     {
-      "epoch": 1.22,
-      "learning_rate": 1.1855019344329057e-05,
-      "loss": 0.1671,
+      "epoch": 0.61,
+      "learning_rate": 1.7559798380937837e-05,
+      "loss": 0.3123,
       "step": 4000
     },
     {
-      "epoch": 1.37,
-      "learning_rate": 1.083689676237019e-05,
-      "loss": 0.1613,
+      "epoch": 0.69,
+      "learning_rate": 1.7254314953413777e-05,
+      "loss": 0.3114,
       "step": 4500
     },
     {
-      "epoch": 1.53,
-      "learning_rate": 9.818774180411322e-06,
-      "loss": 0.1637,
+      "epoch": 0.76,
+      "learning_rate": 1.694883152588972e-05,
+      "loss": 0.2938,
       "step": 5000
     },
     {
-      "epoch": 1.68,
-      "learning_rate": 8.800651598452455e-06,
-      "loss": 0.1611,
+      "epoch": 0.84,
+      "learning_rate": 1.6643348098365667e-05,
+      "loss": 0.297,
       "step": 5500
     },
     {
-      "epoch": 1.83,
-      "learning_rate": 7.782529016493586e-06,
-      "loss": 0.1605,
+      "epoch": 0.92,
+      "learning_rate": 1.6337864670841607e-05,
+      "loss": 0.2818,
       "step": 6000
     },
     {
-      "epoch": 1.99,
-      "learning_rate": 6.764406434534719e-06,
-      "loss": 0.1518,
+      "epoch": 0.99,
+      "learning_rate": 1.60329922101726e-05,
+      "loss": 0.2854,
       "step": 6500
     },
     {
-      "epoch": 2.14,
-      "learning_rate": 5.74628385257585e-06,
-      "loss": 0.0873,
+      "epoch": 1.07,
+      "learning_rate": 1.5727508782648543e-05,
+      "loss": 0.2273,
       "step": 7000
     },
     {
-      "epoch": 2.29,
-      "learning_rate": 4.728161270616982e-06,
-      "loss": 0.0907,
+      "epoch": 1.15,
+      "learning_rate": 1.5422025355124486e-05,
+      "loss": 0.2193,
       "step": 7500
     },
     {
-      "epoch": 2.44,
-      "learning_rate": 3.7100386886581147e-06,
-      "loss": 0.0762,
+      "epoch": 1.22,
+      "learning_rate": 1.511654192760043e-05,
+      "loss": 0.2276,
       "step": 8000
     },
     {
-      "epoch": 2.6,
-      "learning_rate": 2.6919161066992467e-06,
-      "loss": 0.0822,
+      "epoch": 1.3,
+      "learning_rate": 1.4811058500076371e-05,
+      "loss": 0.2151,
       "step": 8500
     },
     {
-      "epoch": 2.75,
-      "learning_rate": 1.6737935247403788e-06,
-      "loss": 0.0887,
+      "epoch": 1.37,
+      "learning_rate": 1.4505575072552316e-05,
+      "loss": 0.2193,
       "step": 9000
     },
     {
-      "epoch": 2.9,
-      "learning_rate": 6.55670942781511e-07,
-      "loss": 0.0868,
+      "epoch": 1.45,
+      "learning_rate": 1.4200702611883306e-05,
+      "loss": 0.2363,
       "step": 9500
     },
     {
-      "epoch": 3.0,
-      "step": 9822,
-      "total_flos": 7.320996509179853e+16,
-      "train_loss": 0.18336697417057063,
-      "train_runtime": 5113.4763,
-      "train_samples_per_second": 61.451,
-      "train_steps_per_second": 1.921
+      "epoch": 1.53,
+      "learning_rate": 1.3897052084924394e-05,
+      "loss": 0.2273,
+      "step": 10000
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.3591568657400339e-05,
+      "loss": 0.2275,
+      "step": 10500
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.328608522987628e-05,
+      "loss": 0.2247,
+      "step": 11000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.2980601802352224e-05,
+      "loss": 0.2133,
+      "step": 11500
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 1.2675118374828165e-05,
+      "loss": 0.2222,
+      "step": 12000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 1.236963494730411e-05,
+      "loss": 0.2136,
+      "step": 12500
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.2064151519780052e-05,
+      "loss": 0.2185,
+      "step": 13000
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.1758668092255997e-05,
+      "loss": 0.1531,
+      "step": 13500
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 1.145318466473194e-05,
+      "loss": 0.1496,
+      "step": 14000
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 1.1147701237207882e-05,
+      "loss": 0.1467,
+      "step": 14500
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.0842217809683826e-05,
+      "loss": 0.1674,
+      "step": 15000
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 1.0536734382159768e-05,
+      "loss": 0.1505,
+      "step": 15500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.0231250954635711e-05,
+      "loss": 0.1447,
+      "step": 16000
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 9.925767527111655e-06,
+      "loss": 0.1407,
+      "step": 16500
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 9.620284099587598e-06,
+      "loss": 0.153,
+      "step": 17000
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 9.314800672063541e-06,
+      "loss": 0.1534,
+      "step": 17500
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 9.009317244539484e-06,
+      "loss": 0.1499,
+      "step": 18000
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 8.703833817015428e-06,
+      "loss": 0.1605,
+      "step": 18500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 8.398350389491371e-06,
+      "loss": 0.1674,
+      "step": 19000
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 8.093477928822362e-06,
+      "loss": 0.1524,
+      "step": 19500
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 7.788605468153352e-06,
+      "loss": 0.0993,
+      "step": 20000
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 7.483122040629296e-06,
+      "loss": 0.081,
+      "step": 20500
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 7.17763861310524e-06,
+      "loss": 0.0811,
+      "step": 21000
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 6.872155185581183e-06,
+      "loss": 0.0825,
+      "step": 21500
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 6.566671758057126e-06,
+      "loss": 0.0823,
+      "step": 22000
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 6.261188330533069e-06,
+      "loss": 0.0808,
+      "step": 22500
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 5.955704903009012e-06,
+      "loss": 0.078,
+      "step": 23000
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 5.650221475484955e-06,
+      "loss": 0.0764,
+      "step": 23500
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 5.344738047960899e-06,
+      "loss": 0.0769,
+      "step": 24000
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 5.040476554146938e-06,
+      "loss": 0.0861,
+      "step": 24500
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 4.734993126622881e-06,
+      "loss": 0.0956,
+      "step": 25000
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 4.429509699098824e-06,
+      "loss": 0.0827,
+      "step": 25500
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 4.124026271574768e-06,
+      "loss": 0.0836,
+      "step": 26000
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 3.81854284405071e-06,
+      "loss": 0.0529,
+      "step": 26500
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 3.5130594165266534e-06,
+      "loss": 0.0371,
+      "step": 27000
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 3.208797922712693e-06,
+      "loss": 0.0408,
+      "step": 27500
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 2.903314495188636e-06,
+      "loss": 0.044,
+      "step": 28000
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 2.5978310676645796e-06,
+      "loss": 0.034,
+      "step": 28500
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 2.2923476401405224e-06,
+      "loss": 0.041,
+      "step": 29000
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 1.9868642126164657e-06,
+      "loss": 0.0402,
+      "step": 29500
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 1.6813807850924087e-06,
+      "loss": 0.0321,
+      "step": 30000
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 1.375897357568352e-06,
+      "loss": 0.0349,
+      "step": 30500
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 1.0704139300442952e-06,
+      "loss": 0.0422,
+      "step": 31000
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 7.649305025202383e-07,
+      "loss": 0.041,
+      "step": 31500
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 4.594470749961815e-07,
+      "loss": 0.0445,
+      "step": 32000
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 1.5396364747212466e-07,
+      "loss": 0.0376,
+      "step": 32500
+    },
+    {
+      "epoch": 5.0,
+      "step": 32735,
+      "total_flos": 1.2201660848633088e+17,
+      "train_loss": 0.16656628906298943,
+      "train_runtime": 2957.7214,
+      "train_samples_per_second": 177.067,
+      "train_steps_per_second": 11.068
     }
   ],
-  "max_steps": 9822,
-  "num_train_epochs": 3,
-  "total_flos": 7.320996509179853e+16,
+  "max_steps": 32735,
+  "num_train_epochs": 5,
+  "total_flos": 1.2201660848633088e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d4cad0cb38dc1cfd9d71e44429d449ab3e51d66d08c9d80a3c795b07ae099297
+oid sha256:eccc8be0fe7ae8f9eab422b35124ed3dbb2c8f9a6c5ad67ccb64c02713b3cadd
 size 3963