Training in progress, epoch 2

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26cd8608c74765cf0d7e63a3bd73516440951ea66a7798b1db8fabaf293cd73d
+oid sha256:934a427b00be68da3ed8fde7b1bcdd15d4eee49375d3c550286d6c6b91d75482
 size 267832560

run-2/checkpoint-4276/config.json

@@ -20,6 +20,6 @@
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.35.2",
+  "transformers_version": "4.38.2",
   "vocab_size": 30522
 }

run-2/checkpoint-4276/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2641a0fb5e8098c1e4392dd7fb772522aab8196dfa550989e9b11678c3074660
+oid sha256:934a427b00be68da3ed8fde7b1bcdd15d4eee49375d3c550286d6c6b91d75482
 size 267832560

run-2/checkpoint-4276/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73bb318ea7868b4d3b5448eee3af2735d8e81824613de6e44e76ff1f949a34cf
+oid sha256:2f41bf4a3c1d763cdaef986d5e50ce022d06df3a30a5eab7fc97e283653bcb0f
 size 535727290

run-2/checkpoint-4276/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff56f6cf771ddd7d20265eca53b8a64af4c290f26007b82e13456dec8a4076f6
-size 14308
+oid sha256:00e672a9362c6f5f488b05417701d476fdfce4dbe9c5c1eff5573f6a228e6183
+size 14244

run-2/checkpoint-4276/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c001a95dc22efb46374f1f965702e4e20dbf9fde40fb4980df8a756842d4e0ac
+oid sha256:4d883f9bbfa01924e0f8592716e3b9514c8da859b9f26198cdd40cf5b4791cd9
 size 1064

run-2/checkpoint-4276/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.49329306040311344,
-  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-2/checkpoint-3207",
-  "epoch": 4.0,
+  "best_metric": 0.4691032179514943,
+  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-2/checkpoint-4276",
+  "epoch": 2.0,
   "eval_steps": 500,
   "global_step": 4276,
   "is_hyper_param_search": true,
@@ -9,100 +9,92 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.47,
-      "learning_rate": 5.944660653119031e-06,
-      "loss": 0.5558,
+      "epoch": 0.23,
+      "grad_norm": 4.28505277633667,
+      "learning_rate": 3.0702893894484785e-06,
+      "loss": 0.6069,
       "step": 500
     },
     {
-      "epoch": 0.94,
-      "learning_rate": 5.157496901381872e-06,
-      "loss": 0.5053,
+      "epoch": 0.47,
+      "grad_norm": 9.482794761657715,
+      "learning_rate": 2.9196373094951675e-06,
+      "loss": 0.5628,
       "step": 1000
     },
     {
-      "epoch": 1.0,
-      "eval_loss": 0.4907635748386383,
-      "eval_matthews_correlation": 0.42854888313294803,
-      "eval_runtime": 0.8054,
-      "eval_samples_per_second": 1295.012,
-      "eval_steps_per_second": 81.947,
-      "step": 1069
-    },
-    {
-      "epoch": 1.4,
-      "learning_rate": 4.370333149644712e-06,
-      "loss": 0.4286,
+      "epoch": 0.7,
+      "grad_norm": 22.521339416503906,
+      "learning_rate": 2.7689852295418565e-06,
+      "loss": 0.5565,
       "step": 1500
     },
     {
-      "epoch": 1.87,
-      "learning_rate": 3.5831693979075514e-06,
-      "loss": 0.3992,
+      "epoch": 0.94,
+      "grad_norm": 26.7753849029541,
+      "learning_rate": 2.6183331495885454e-06,
+      "loss": 0.5184,
       "step": 2000
     },
     {
-      "epoch": 2.0,
-      "eval_loss": 0.5015696883201599,
-      "eval_matthews_correlation": 0.48409375114357234,
-      "eval_runtime": 0.8234,
-      "eval_samples_per_second": 1266.73,
-      "eval_steps_per_second": 80.157,
+      "epoch": 1.0,
+      "eval_loss": 0.5730993747711182,
+      "eval_matthews_correlation": 0.3853198145814999,
+      "eval_runtime": 0.7612,
+      "eval_samples_per_second": 1370.225,
+      "eval_steps_per_second": 86.706,
       "step": 2138
     },
     {
-      "epoch": 2.34,
-      "learning_rate": 2.796005646170392e-06,
-      "loss": 0.3434,
+      "epoch": 1.17,
+      "grad_norm": 17.77669334411621,
+      "learning_rate": 2.4676810696352344e-06,
+      "loss": 0.4619,
       "step": 2500
     },
     {
-      "epoch": 2.81,
-      "learning_rate": 2.008841894433232e-06,
-      "loss": 0.3371,
+      "epoch": 1.4,
+      "grad_norm": 37.4239387512207,
+      "learning_rate": 2.3170289896819234e-06,
+      "loss": 0.5014,
       "step": 3000
     },
     {
-      "epoch": 3.0,
-      "eval_loss": 0.5880187749862671,
-      "eval_matthews_correlation": 0.49329306040311344,
-      "eval_runtime": 0.9037,
-      "eval_samples_per_second": 1154.15,
-      "eval_steps_per_second": 73.033,
-      "step": 3207
-    },
-    {
-      "epoch": 3.27,
-      "learning_rate": 1.2216781426960722e-06,
-      "loss": 0.2988,
+      "epoch": 1.64,
+      "grad_norm": 46.75569534301758,
+      "learning_rate": 2.1663769097286124e-06,
+      "loss": 0.492,
       "step": 3500
     },
     {
-      "epoch": 3.74,
-      "learning_rate": 4.345143909589123e-07,
-      "loss": 0.2797,
+      "epoch": 1.87,
+      "grad_norm": 66.9134750366211,
+      "learning_rate": 2.0157248297753013e-06,
+      "loss": 0.4809,
       "step": 4000
     },
     {
-      "epoch": 4.0,
-      "eval_loss": 0.6395880579948425,
-      "eval_matthews_correlation": 0.49249265259737396,
-      "eval_runtime": 1.057,
-      "eval_samples_per_second": 986.747,
-      "eval_steps_per_second": 62.44,
+      "epoch": 2.0,
+      "eval_loss": 0.6646500825881958,
+      "eval_matthews_correlation": 0.4691032179514943,
+      "eval_runtime": 0.8224,
+      "eval_samples_per_second": 1268.193,
+      "eval_steps_per_second": 80.25,
       "step": 4276
     }
   ],
   "logging_steps": 500,
-  "max_steps": 4276,
-  "num_train_epochs": 4,
+  "max_steps": 10690,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 150793435338816.0,
+  "total_flos": 65200091402940.0,
+  "train_batch_size": 4,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 6.7318244048561916e-06,
-    "num_train_epochs": 4,
-    "per_device_train_batch_size": 8,
-    "seed": 14
+    "learning_rate": 3.2209414694017896e-06,
+    "num_train_epochs": 5,
+    "per_device_train_batch_size": 4,
+    "seed": 16
   }
 }

run-2/checkpoint-4276/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:618b6f744377ad587ba25f1af0a1c4f702e5743116ab1fe52e9ecc52d00dc59b
-size 4664
+oid sha256:88d2c3d6804ca2d9d22cb74f328c5ae8ec320f8d12a0ef15ea5ae2037f02bd85
+size 4984

runs/Mar16_01-46-29_6e5f088ca464/events.out.tfevents.1710554238.6e5f088ca464.226.4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a2728a30252b8c7698d4b1cbbee9e9398c0d44f7fa3ab038a041c0d6da1827ad
-size 5979
+oid sha256:9e275f4a807e39d73a49dad5d1ea7c25874258e28fe23616f020c0ad00492050
+size 7158