diff --git a/README.md b/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..12c94840fea8a3afce02142e31cf01d60d62dcbf
--- /dev/null
+++ b/README.md
@@ -0,0 +1,57 @@
+---
+license: apache-2.0
+base_model: bert-base-cased
+tags:
+- generated_from_trainer
+model-index:
+- name: bert_trainer
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# bert_trainer
+
+This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the None dataset.
+It achieves the following results on the evaluation set:
+- eval_loss: 0.6901
+- eval_accuracy: 0.7558
+- eval_runtime: 30.0507
+- eval_samples_per_second: 103.292
+- eval_steps_per_second: 25.823
+- epoch: 0.16
+- step: 250
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 6.036609853019458e-06
+- train_batch_size: 8
+- eval_batch_size: 4
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: constant_with_warmup
+- lr_scheduler_warmup_steps: 50
+- training_steps: 3000
+
+### Framework versions
+
+- Transformers 4.36.0.dev0
+- Pytorch 2.1.1+cu121
+- Datasets 2.15.0
+- Tokenizers 0.15.0
diff --git a/config.json b/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/model.safetensors b/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..013d9b0b2d4d6c4aaa4e5b068c6535d11a141575
--- /dev/null
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd32e24b75429e0cb9367040b8e23512009bab779fbedc84c7cd676a547332c8
+size 433279996
diff --git a/run-0/checkpoint-2500/config.json b/run-0/checkpoint-2500/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-0/checkpoint-2500/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-0/checkpoint-2500/model.safetensors b/run-0/checkpoint-2500/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d9b70e6916f1aba89cd38259d579ec3a23c0826e
--- /dev/null
+++ b/run-0/checkpoint-2500/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d3e6ee528249c0e1c6388d2e72c4e47db8e42f96022ddc8f107119bbde5e135
+size 433279996
diff --git a/run-0/checkpoint-2500/optimizer.pt b/run-0/checkpoint-2500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..195d768201a726448ec3ad142d2dc8f6fdd7a37b
--- /dev/null
+++ b/run-0/checkpoint-2500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48b0eabf74e09755cd69eb9a84a842e5681649a25f2c588f1e78331b8358fd78
+size 866681082
diff --git a/run-0/checkpoint-2500/rng_state.pth b/run-0/checkpoint-2500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b68d4eaf43291f178fb309f1adb0a7b459dd867c
--- /dev/null
+++ b/run-0/checkpoint-2500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e254f06883fec1e30ea9be59e97a84e2814064af09a93a2e0e9f4655ea082cd
+size 14244
diff --git a/run-0/checkpoint-2500/scheduler.pt b/run-0/checkpoint-2500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6afcc1dd13ee28c42085442c72968a45793dc239
--- /dev/null
+++ b/run-0/checkpoint-2500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06617cdbcc181bf9ec7d4df7aa9e71ef685d4759e304bde8f28a58df45e09177
+size 1064
diff --git a/run-0/checkpoint-2500/trainer_state.json b/run-0/checkpoint-2500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..0d5522cf8856455f0fa3be6d54a1e1b81449c798
--- /dev/null
+++ b/run-0/checkpoint-2500/trainer_state.json
@@ -0,0 +1,714 @@
+{
+  "best_metric": 0.8743556701030928,
+  "best_model_checkpoint": "bert_trainer/run-0/checkpoint-1750",
+  "epoch": 3.2216494845360826,
+  "eval_steps": 250,
+  "global_step": 2500,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.650169893369787e-06,
+      "loss": 1.4561,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 1.2146,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 1.0949,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.9846,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.9161,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.8466,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.7954,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.7311,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6688,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6816,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.7857603092783505,
+      "eval_loss": 0.6313713192939758,
+      "eval_runtime": 78.939,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 9.83,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6339,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6102,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6572,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6575,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5946,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5779,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5773,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5461,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.576,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4864,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8176546391752577,
+      "eval_loss": 0.5011568069458008,
+      "eval_runtime": 78.7987,
+      "eval_samples_per_second": 39.392,
+      "eval_steps_per_second": 9.848,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5618,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5325,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4656,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4966,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.457,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4715,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4621,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4679,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4236,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.399,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8443943298969072,
+      "eval_loss": 0.4396892488002777,
+      "eval_runtime": 78.9384,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 9.83,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4321,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3953,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.379,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3415,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3705,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3612,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4108,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4167,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3916,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3947,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8579252577319587,
+      "eval_loss": 0.40539947152137756,
+      "eval_runtime": 78.7689,
+      "eval_samples_per_second": 39.406,
+      "eval_steps_per_second": 9.852,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3812,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3804,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3783,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.337,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3191,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3277,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3601,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3932,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3964,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3705,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8524484536082474,
+      "eval_loss": 0.40534982085227966,
+      "eval_runtime": 78.953,
+      "eval_samples_per_second": 39.315,
+      "eval_steps_per_second": 9.829,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4567,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.384,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3205,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3841,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.322,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3742,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3407,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3342,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3068,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3288,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8688788659793815,
+      "eval_loss": 0.36364319920539856,
+      "eval_runtime": 78.9463,
+      "eval_samples_per_second": 39.318,
+      "eval_steps_per_second": 9.829,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2918,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3571,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3341,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3162,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2384,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2934,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2686,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2806,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2897,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2417,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8743556701030928,
+      "eval_loss": 0.39001935720443726,
+      "eval_runtime": 78.9576,
+      "eval_samples_per_second": 39.312,
+      "eval_steps_per_second": 9.828,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2849,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2653,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2266,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2401,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2208,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2615,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2913,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2474,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2388,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2374,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8730670103092784,
+      "eval_loss": 0.39385583996772766,
+      "eval_runtime": 78.9575,
+      "eval_samples_per_second": 39.312,
+      "eval_steps_per_second": 9.828,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2697,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2592,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2636,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2349,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2697,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2507,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2608,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2463,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2408,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3525,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8627577319587629,
+      "eval_loss": 0.4113520085811615,
+      "eval_runtime": 78.9601,
+      "eval_samples_per_second": 39.311,
+      "eval_steps_per_second": 9.828,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3033,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2186,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2619,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1473,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1662,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1863,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1869,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1729,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1998,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1778,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.873389175257732,
+      "eval_loss": 0.44405418634414673,
+      "eval_runtime": 78.9751,
+      "eval_samples_per_second": 39.304,
+      "eval_steps_per_second": 9.826,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.0521042044565504e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 5.300339786739574e-06,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-0/checkpoint-2500/training_args.bin b/run-0/checkpoint-2500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..ed612abf640aadfc105f23e03656a70e24f64012
--- /dev/null
+++ b/run-0/checkpoint-2500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff966c9d210a3dcc91765468d3bd580c3a94cec8f318bfe2ceb18c02bfa0afcf
+size 4728
diff --git a/run-0/checkpoint-2750/config.json b/run-0/checkpoint-2750/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-0/checkpoint-2750/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-0/checkpoint-2750/model.safetensors b/run-0/checkpoint-2750/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d85fb526b5aead523a5530042f0541c4ab977949
--- /dev/null
+++ b/run-0/checkpoint-2750/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1336ad1a77da09cbf25876d022a58770130c3b2ffe4b80c420562bcdc7d1ed9d
+size 433279996
diff --git a/run-0/checkpoint-2750/optimizer.pt b/run-0/checkpoint-2750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1a7af757084b4f22e993513dbac5aef7413da7e3
--- /dev/null
+++ b/run-0/checkpoint-2750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1162a3ddf040cd09a94a4dd7463544f40b0659e8996a1d72eb0f3668b66b34d1
+size 866681082
diff --git a/run-0/checkpoint-2750/rng_state.pth b/run-0/checkpoint-2750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..eccfea317f0c1eb50423f8ebdfd26baba83a270c
--- /dev/null
+++ b/run-0/checkpoint-2750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2f146112a2b6651d5740f314c7a7760a6b10b307030d3ca8f929af57ce20006
+size 14244
diff --git a/run-0/checkpoint-2750/scheduler.pt b/run-0/checkpoint-2750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..df4bf57ce0cc683340536da88e18075ff404b462
--- /dev/null
+++ b/run-0/checkpoint-2750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2990860678a4cf60168b8269ff8709a95193f5ee58d391b8ba0badb27d72ce0
+size 1064
diff --git a/run-0/checkpoint-2750/trainer_state.json b/run-0/checkpoint-2750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..2b3cdb43c236983788c7e41785c396bd5fb79795
--- /dev/null
+++ b/run-0/checkpoint-2750/trainer_state.json
@@ -0,0 +1,783 @@
+{
+  "best_metric": 0.8743556701030928,
+  "best_model_checkpoint": "bert_trainer/run-0/checkpoint-1750",
+  "epoch": 3.5438144329896906,
+  "eval_steps": 250,
+  "global_step": 2750,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.650169893369787e-06,
+      "loss": 1.4561,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 1.2146,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 1.0949,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.9846,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.9161,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.8466,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.7954,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.7311,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6688,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6816,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.7857603092783505,
+      "eval_loss": 0.6313713192939758,
+      "eval_runtime": 78.939,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 9.83,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6339,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6102,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6572,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6575,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5946,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5779,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5773,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5461,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.576,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4864,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8176546391752577,
+      "eval_loss": 0.5011568069458008,
+      "eval_runtime": 78.7987,
+      "eval_samples_per_second": 39.392,
+      "eval_steps_per_second": 9.848,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5618,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5325,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4656,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4966,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.457,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4715,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4621,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4679,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4236,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.399,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8443943298969072,
+      "eval_loss": 0.4396892488002777,
+      "eval_runtime": 78.9384,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 9.83,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4321,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3953,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.379,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3415,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3705,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3612,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4108,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4167,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3916,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3947,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8579252577319587,
+      "eval_loss": 0.40539947152137756,
+      "eval_runtime": 78.7689,
+      "eval_samples_per_second": 39.406,
+      "eval_steps_per_second": 9.852,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3812,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3804,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3783,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.337,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3191,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3277,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3601,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3932,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3964,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3705,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8524484536082474,
+      "eval_loss": 0.40534982085227966,
+      "eval_runtime": 78.953,
+      "eval_samples_per_second": 39.315,
+      "eval_steps_per_second": 9.829,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4567,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.384,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3205,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3841,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.322,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3742,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3407,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3342,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3068,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3288,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8688788659793815,
+      "eval_loss": 0.36364319920539856,
+      "eval_runtime": 78.9463,
+      "eval_samples_per_second": 39.318,
+      "eval_steps_per_second": 9.829,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2918,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3571,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3341,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3162,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2384,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2934,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2686,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2806,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2897,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2417,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8743556701030928,
+      "eval_loss": 0.39001935720443726,
+      "eval_runtime": 78.9576,
+      "eval_samples_per_second": 39.312,
+      "eval_steps_per_second": 9.828,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2849,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2653,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2266,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2401,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2208,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2615,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2913,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2474,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2388,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2374,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8730670103092784,
+      "eval_loss": 0.39385583996772766,
+      "eval_runtime": 78.9575,
+      "eval_samples_per_second": 39.312,
+      "eval_steps_per_second": 9.828,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2697,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2592,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2636,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2349,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2697,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2507,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2608,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2463,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2408,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3525,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8627577319587629,
+      "eval_loss": 0.4113520085811615,
+      "eval_runtime": 78.9601,
+      "eval_samples_per_second": 39.311,
+      "eval_steps_per_second": 9.828,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3033,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2186,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2619,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1473,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1662,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1863,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1869,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1729,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1998,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1778,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.873389175257732,
+      "eval_loss": 0.44405418634414673,
+      "eval_runtime": 78.9751,
+      "eval_samples_per_second": 39.304,
+      "eval_steps_per_second": 9.826,
+      "step": 2500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.171,
+      "step": 2525
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1642,
+      "step": 2550
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2303,
+      "step": 2575
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2194,
+      "step": 2600
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1756,
+      "step": 2625
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1956,
+      "step": 2650
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2669,
+      "step": 2675
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1514,
+      "step": 2700
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1263,
+      "step": 2725
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.259,
+      "step": 2750
+    },
+    {
+      "epoch": 3.54,
+      "eval_accuracy": 0.8566365979381443,
+      "eval_loss": 0.4988997280597687,
+      "eval_runtime": 78.9372,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 9.831,
+      "step": 2750
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.1573514614421504e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 5.300339786739574e-06,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-0/checkpoint-2750/training_args.bin b/run-0/checkpoint-2750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..ed612abf640aadfc105f23e03656a70e24f64012
--- /dev/null
+++ b/run-0/checkpoint-2750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff966c9d210a3dcc91765468d3bd580c3a94cec8f318bfe2ceb18c02bfa0afcf
+size 4728
diff --git a/run-0/checkpoint-3000/config.json b/run-0/checkpoint-3000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-0/checkpoint-3000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-0/checkpoint-3000/model.safetensors b/run-0/checkpoint-3000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e6b3751a07b0e06aa0faa00b5fb73a1de0138181
--- /dev/null
+++ b/run-0/checkpoint-3000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97635f937f6839acf8375df2ae47c49059b50b550f9a87c413496a140714f269
+size 433279996
diff --git a/run-0/checkpoint-3000/optimizer.pt b/run-0/checkpoint-3000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1bdda14e4ec268063efeb2ae15e1100ced347770
--- /dev/null
+++ b/run-0/checkpoint-3000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:255f8e93744d2bc7a27bc7622955b2b19abef2409dcd84b03f15ac2f69afffbf
+size 866681082
diff --git a/run-0/checkpoint-3000/rng_state.pth b/run-0/checkpoint-3000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..134d7b84b2bcb7c8af8198d9d691b9391bfa81d0
--- /dev/null
+++ b/run-0/checkpoint-3000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5475dcc749e5bb84d23e503506ffb30e2c214153e88098ffa2eba5466de55a64
+size 14244
diff --git a/run-0/checkpoint-3000/scheduler.pt b/run-0/checkpoint-3000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..121c91960c47c9b460c2730b0ddb66eecd67997f
--- /dev/null
+++ b/run-0/checkpoint-3000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11b18eab96c7f27ea28cf71b3d2f6754e002f163c751b2006fe6b5fd2f4b1877
+size 1064
diff --git a/run-0/checkpoint-3000/trainer_state.json b/run-0/checkpoint-3000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..db4d1128851b98c658dde2580e09dd609153bc53
--- /dev/null
+++ b/run-0/checkpoint-3000/trainer_state.json
@@ -0,0 +1,852 @@
+{
+  "best_metric": 0.8798324742268041,
+  "best_model_checkpoint": "bert_trainer/run-0/checkpoint-3000",
+  "epoch": 3.865979381443299,
+  "eval_steps": 250,
+  "global_step": 3000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.650169893369787e-06,
+      "loss": 1.4561,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 1.2146,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 1.0949,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.9846,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.9161,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.8466,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.7954,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.7311,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6688,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6816,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.7857603092783505,
+      "eval_loss": 0.6313713192939758,
+      "eval_runtime": 78.939,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 9.83,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6339,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6102,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6572,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.6575,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5946,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5779,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5773,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5461,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.576,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4864,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8176546391752577,
+      "eval_loss": 0.5011568069458008,
+      "eval_runtime": 78.7987,
+      "eval_samples_per_second": 39.392,
+      "eval_steps_per_second": 9.848,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5618,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.5325,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4656,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4966,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.457,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4715,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4621,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4679,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4236,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.399,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8443943298969072,
+      "eval_loss": 0.4396892488002777,
+      "eval_runtime": 78.9384,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 9.83,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4321,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3953,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.379,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3415,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3705,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3612,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4108,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4167,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3916,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3947,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8579252577319587,
+      "eval_loss": 0.40539947152137756,
+      "eval_runtime": 78.7689,
+      "eval_samples_per_second": 39.406,
+      "eval_steps_per_second": 9.852,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3812,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3804,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3783,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.337,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3191,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3277,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3601,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3932,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3964,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3705,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8524484536082474,
+      "eval_loss": 0.40534982085227966,
+      "eval_runtime": 78.953,
+      "eval_samples_per_second": 39.315,
+      "eval_steps_per_second": 9.829,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.4567,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.384,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3205,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3841,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.322,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3742,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3407,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3342,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3068,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3288,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8688788659793815,
+      "eval_loss": 0.36364319920539856,
+      "eval_runtime": 78.9463,
+      "eval_samples_per_second": 39.318,
+      "eval_steps_per_second": 9.829,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2918,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3571,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3341,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3162,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2384,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2934,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2686,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2806,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2897,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2417,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8743556701030928,
+      "eval_loss": 0.39001935720443726,
+      "eval_runtime": 78.9576,
+      "eval_samples_per_second": 39.312,
+      "eval_steps_per_second": 9.828,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2849,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2653,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2266,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2401,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2208,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2615,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2913,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2474,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2388,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2374,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8730670103092784,
+      "eval_loss": 0.39385583996772766,
+      "eval_runtime": 78.9575,
+      "eval_samples_per_second": 39.312,
+      "eval_steps_per_second": 9.828,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2697,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2592,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2636,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2349,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2697,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2507,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2608,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2463,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2408,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3525,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8627577319587629,
+      "eval_loss": 0.4113520085811615,
+      "eval_runtime": 78.9601,
+      "eval_samples_per_second": 39.311,
+      "eval_steps_per_second": 9.828,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3033,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2186,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2619,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1473,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1662,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1863,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1869,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1729,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1998,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1778,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.873389175257732,
+      "eval_loss": 0.44405418634414673,
+      "eval_runtime": 78.9751,
+      "eval_samples_per_second": 39.304,
+      "eval_steps_per_second": 9.826,
+      "step": 2500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.171,
+      "step": 2525
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1642,
+      "step": 2550
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2303,
+      "step": 2575
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2194,
+      "step": 2600
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1756,
+      "step": 2625
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1956,
+      "step": 2650
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2669,
+      "step": 2675
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1514,
+      "step": 2700
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1263,
+      "step": 2725
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.259,
+      "step": 2750
+    },
+    {
+      "epoch": 3.54,
+      "eval_accuracy": 0.8566365979381443,
+      "eval_loss": 0.4988997280597687,
+      "eval_runtime": 78.9372,
+      "eval_samples_per_second": 39.322,
+      "eval_steps_per_second": 9.831,
+      "step": 2750
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1378,
+      "step": 2775
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1588,
+      "step": 2800
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.179,
+      "step": 2825
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.3413,
+      "step": 2850
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.169,
+      "step": 2875
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.208,
+      "step": 2900
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1602,
+      "step": 2925
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1406,
+      "step": 2950
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.1519,
+      "step": 2975
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 5.300339786739574e-06,
+      "loss": 0.2028,
+      "step": 3000
+    },
+    {
+      "epoch": 3.87,
+      "eval_accuracy": 0.8798324742268041,
+      "eval_loss": 0.46966859698295593,
+      "eval_runtime": 78.9816,
+      "eval_samples_per_second": 39.3,
+      "eval_steps_per_second": 9.825,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.2625987184277504e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 5.300339786739574e-06,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-0/checkpoint-3000/training_args.bin b/run-0/checkpoint-3000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..ed612abf640aadfc105f23e03656a70e24f64012
--- /dev/null
+++ b/run-0/checkpoint-3000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff966c9d210a3dcc91765468d3bd580c3a94cec8f318bfe2ceb18c02bfa0afcf
+size 4728
diff --git a/run-1/checkpoint-2000/config.json b/run-1/checkpoint-2000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-1/checkpoint-2000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-1/checkpoint-2000/model.safetensors b/run-1/checkpoint-2000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a8010001896f2472f492851d74d804defe6e47f0
--- /dev/null
+++ b/run-1/checkpoint-2000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d2979aaf550705ed15db92c754402c39bde2cf239b007351cd8d694f464ea9d
+size 433279996
diff --git a/run-1/checkpoint-2000/optimizer.pt b/run-1/checkpoint-2000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..454aac267edf504cf40bc189b69e25eb7d6695ff
--- /dev/null
+++ b/run-1/checkpoint-2000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:797fb321bed095050f952383f9272d50d76f008a2b10fdb902f0877056a930b7
+size 866681082
diff --git a/run-1/checkpoint-2000/rng_state.pth b/run-1/checkpoint-2000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..46bfd402f6bfeab3339ef84c13f8bea7848fb6b2
--- /dev/null
+++ b/run-1/checkpoint-2000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5e36c9a58997a8fd5116115315a1541147a866ccc21b5fd8f202de7af5118db
+size 14244
diff --git a/run-1/checkpoint-2000/scheduler.pt b/run-1/checkpoint-2000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..47a76f3370088b7968b8bc5c7998f52e0a621e4c
--- /dev/null
+++ b/run-1/checkpoint-2000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd52d0eb8a495b5ec54fb0f307d28eec2eec58a5a5ce1ea1132a7d75494f126c
+size 1064
diff --git a/run-1/checkpoint-2000/trainer_state.json b/run-1/checkpoint-2000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..cfdf261aab16857e5f04f042377c234ef50f67ed
--- /dev/null
+++ b/run-1/checkpoint-2000/trainer_state.json
@@ -0,0 +1,576 @@
+{
+  "best_metric": 0.8833762886597938,
+  "best_model_checkpoint": "bert_trainer/run-1/checkpoint-2000",
+  "epoch": 2.5773195876288657,
+  "eval_steps": 250,
+  "global_step": 2000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 8.555424210708822e-06,
+      "loss": 1.3732,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 1.0422,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.913,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.7626,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.7043,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.6356,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.6031,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5454,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4999,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5111,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8279639175257731,
+      "eval_loss": 0.4562760293483734,
+      "eval_runtime": 65.4867,
+      "eval_samples_per_second": 47.399,
+      "eval_steps_per_second": 11.85,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.511,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4849,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5027,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.565,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4704,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4931,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4647,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4491,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4752,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3899,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8630798969072165,
+      "eval_loss": 0.3841710388660431,
+      "eval_runtime": 65.5356,
+      "eval_samples_per_second": 47.364,
+      "eval_steps_per_second": 11.841,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.477,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3964,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3763,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3928,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4235,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4024,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4054,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3731,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3749,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3315,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8666237113402062,
+      "eval_loss": 0.3664605915546417,
+      "eval_runtime": 65.4939,
+      "eval_samples_per_second": 47.394,
+      "eval_steps_per_second": 11.848,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3572,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2822,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2383,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2685,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3076,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2832,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2976,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2826,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2585,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2816,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8753221649484536,
+      "eval_loss": 0.3534678518772125,
+      "eval_runtime": 65.5301,
+      "eval_samples_per_second": 47.368,
+      "eval_steps_per_second": 11.842,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2583,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2405,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2626,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1829,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.217,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2426,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2937,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.283,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2939,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2477,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8663015463917526,
+      "eval_loss": 0.37554699182510376,
+      "eval_runtime": 65.5483,
+      "eval_samples_per_second": 47.354,
+      "eval_steps_per_second": 11.839,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3323,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2489,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2451,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2911,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2348,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2565,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1898,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2489,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2442,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2745,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8717783505154639,
+      "eval_loss": 0.3890085220336914,
+      "eval_runtime": 65.5564,
+      "eval_samples_per_second": 47.349,
+      "eval_steps_per_second": 11.837,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.208,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2804,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.186,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.206,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1658,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1576,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1845,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1601,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1397,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.109,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8814432989690721,
+      "eval_loss": 0.5400919318199158,
+      "eval_runtime": 65.5029,
+      "eval_samples_per_second": 47.387,
+      "eval_steps_per_second": 11.847,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1069,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1495,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1209,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.134,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1357,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1114,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1436,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1775,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1232,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.115,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8833762886597938,
+      "eval_loss": 0.5348294973373413,
+      "eval_runtime": 67.3607,
+      "eval_samples_per_second": 46.08,
+      "eval_steps_per_second": 11.52,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 8414518195998720.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.7110848421417645e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-1/checkpoint-2000/training_args.bin b/run-1/checkpoint-2000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..109d901eeee674be35ec2b799f5af2df7bfd73e8
--- /dev/null
+++ b/run-1/checkpoint-2000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ff7ff5e497d2fd17b0b0a335d1e1e52fdf9ab64c61fc776fe6063c480e0a9f6
+size 4728
diff --git a/run-1/checkpoint-2750/config.json b/run-1/checkpoint-2750/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-1/checkpoint-2750/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-1/checkpoint-2750/model.safetensors b/run-1/checkpoint-2750/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..bd2f5acc4342501ad7f1398d2491b34d590db703
--- /dev/null
+++ b/run-1/checkpoint-2750/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fdf1daffe1979f97111b495ac9166af3936db88484feea619daefa7ee4afe6e
+size 433279996
diff --git a/run-1/checkpoint-2750/optimizer.pt b/run-1/checkpoint-2750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..395c3dcd9cfba1fc35887ad04a1193a52c5fafd4
--- /dev/null
+++ b/run-1/checkpoint-2750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f352afcb5855371e75fc5f7c2a50086a26116ec8440ae8fb5ea58867ecb0b84f
+size 866681082
diff --git a/run-1/checkpoint-2750/rng_state.pth b/run-1/checkpoint-2750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..eccfea317f0c1eb50423f8ebdfd26baba83a270c
--- /dev/null
+++ b/run-1/checkpoint-2750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2f146112a2b6651d5740f314c7a7760a6b10b307030d3ca8f929af57ce20006
+size 14244
diff --git a/run-1/checkpoint-2750/scheduler.pt b/run-1/checkpoint-2750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0235d49c526c54a32729669b75557f34aa31afb1
--- /dev/null
+++ b/run-1/checkpoint-2750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35c90518991fb03cf3fc2f6874522e66785a9b1bb9f1028bbfcb60f66f6de8fd
+size 1064
diff --git a/run-1/checkpoint-2750/trainer_state.json b/run-1/checkpoint-2750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..6d7c6f47ad945c6fcf406c39e94ad446e520d58d
--- /dev/null
+++ b/run-1/checkpoint-2750/trainer_state.json
@@ -0,0 +1,783 @@
+{
+  "best_metric": 0.8833762886597938,
+  "best_model_checkpoint": "bert_trainer/run-1/checkpoint-2000",
+  "epoch": 3.5438144329896906,
+  "eval_steps": 250,
+  "global_step": 2750,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 8.555424210708822e-06,
+      "loss": 1.3732,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 1.0422,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.913,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.7626,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.7043,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.6356,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.6031,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5454,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4999,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5111,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8279639175257731,
+      "eval_loss": 0.4562760293483734,
+      "eval_runtime": 65.4867,
+      "eval_samples_per_second": 47.399,
+      "eval_steps_per_second": 11.85,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.511,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4849,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5027,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.565,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4704,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4931,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4647,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4491,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4752,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3899,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8630798969072165,
+      "eval_loss": 0.3841710388660431,
+      "eval_runtime": 65.5356,
+      "eval_samples_per_second": 47.364,
+      "eval_steps_per_second": 11.841,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.477,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3964,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3763,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3928,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4235,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4024,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4054,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3731,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3749,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3315,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8666237113402062,
+      "eval_loss": 0.3664605915546417,
+      "eval_runtime": 65.4939,
+      "eval_samples_per_second": 47.394,
+      "eval_steps_per_second": 11.848,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3572,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2822,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2383,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2685,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3076,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2832,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2976,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2826,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2585,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2816,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8753221649484536,
+      "eval_loss": 0.3534678518772125,
+      "eval_runtime": 65.5301,
+      "eval_samples_per_second": 47.368,
+      "eval_steps_per_second": 11.842,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2583,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2405,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2626,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1829,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.217,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2426,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2937,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.283,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2939,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2477,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8663015463917526,
+      "eval_loss": 0.37554699182510376,
+      "eval_runtime": 65.5483,
+      "eval_samples_per_second": 47.354,
+      "eval_steps_per_second": 11.839,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3323,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2489,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2451,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2911,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2348,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2565,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1898,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2489,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2442,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2745,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8717783505154639,
+      "eval_loss": 0.3890085220336914,
+      "eval_runtime": 65.5564,
+      "eval_samples_per_second": 47.349,
+      "eval_steps_per_second": 11.837,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.208,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2804,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.186,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.206,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1658,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1576,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1845,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1601,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1397,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.109,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8814432989690721,
+      "eval_loss": 0.5400919318199158,
+      "eval_runtime": 65.5029,
+      "eval_samples_per_second": 47.387,
+      "eval_steps_per_second": 11.847,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1069,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1495,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1209,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.134,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1357,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1114,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1436,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1775,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1232,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.115,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8833762886597938,
+      "eval_loss": 0.5348294973373413,
+      "eval_runtime": 67.3607,
+      "eval_samples_per_second": 46.08,
+      "eval_steps_per_second": 11.52,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1429,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1612,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1565,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1188,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1576,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1832,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1643,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1668,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1984,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1839,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8424613402061856,
+      "eval_loss": 0.6889757513999939,
+      "eval_runtime": 65.603,
+      "eval_samples_per_second": 47.315,
+      "eval_steps_per_second": 11.829,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2497,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1096,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.157,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0431,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0428,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1085,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0781,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1603,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0825,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0855,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.8759664948453608,
+      "eval_loss": 0.6595790982246399,
+      "eval_runtime": 65.5897,
+      "eval_samples_per_second": 47.324,
+      "eval_steps_per_second": 11.831,
+      "step": 2500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0614,
+      "step": 2525
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0686,
+      "step": 2550
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0716,
+      "step": 2575
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1359,
+      "step": 2600
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0863,
+      "step": 2625
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0915,
+      "step": 2650
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0595,
+      "step": 2675
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0714,
+      "step": 2700
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0914,
+      "step": 2725
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0774,
+      "step": 2750
+    },
+    {
+      "epoch": 3.54,
+      "eval_accuracy": 0.8740335051546392,
+      "eval_loss": 0.7505560517311096,
+      "eval_runtime": 65.5758,
+      "eval_samples_per_second": 47.335,
+      "eval_steps_per_second": 11.834,
+      "step": 2750
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.156930472414208e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.7110848421417645e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-1/checkpoint-2750/training_args.bin b/run-1/checkpoint-2750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..109d901eeee674be35ec2b799f5af2df7bfd73e8
--- /dev/null
+++ b/run-1/checkpoint-2750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ff7ff5e497d2fd17b0b0a335d1e1e52fdf9ab64c61fc776fe6063c480e0a9f6
+size 4728
diff --git a/run-1/checkpoint-3000/config.json b/run-1/checkpoint-3000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-1/checkpoint-3000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-1/checkpoint-3000/model.safetensors b/run-1/checkpoint-3000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..34f83183cbe147b3c9b8c42b78a397a1f4f02a21
--- /dev/null
+++ b/run-1/checkpoint-3000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de060276be3bcfe668ca25c8af00e40cad29ee8aaa2f3b02863a1935940a71e1
+size 433279996
diff --git a/run-1/checkpoint-3000/optimizer.pt b/run-1/checkpoint-3000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..82782b2d86815a1b6a1276cd29cc4f44e1de8313
--- /dev/null
+++ b/run-1/checkpoint-3000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10522d5e8817f1178a99dd7c1589a4fabda59fd8a020b8326b29d1d247dd0cdf
+size 866681082
diff --git a/run-1/checkpoint-3000/rng_state.pth b/run-1/checkpoint-3000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..134d7b84b2bcb7c8af8198d9d691b9391bfa81d0
--- /dev/null
+++ b/run-1/checkpoint-3000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5475dcc749e5bb84d23e503506ffb30e2c214153e88098ffa2eba5466de55a64
+size 14244
diff --git a/run-1/checkpoint-3000/scheduler.pt b/run-1/checkpoint-3000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d8455aa1444c2f1f85c9fe850d966f461bfc0cd7
--- /dev/null
+++ b/run-1/checkpoint-3000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80bef91821d2e324748694916761e06c6958f568b9d748689db5bb337914680c
+size 1064
diff --git a/run-1/checkpoint-3000/trainer_state.json b/run-1/checkpoint-3000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c2ca14954ab2c1ac8accb0e34759bb98a1beaff
--- /dev/null
+++ b/run-1/checkpoint-3000/trainer_state.json
@@ -0,0 +1,852 @@
+{
+  "best_metric": 0.8833762886597938,
+  "best_model_checkpoint": "bert_trainer/run-1/checkpoint-2000",
+  "epoch": 3.865979381443299,
+  "eval_steps": 250,
+  "global_step": 3000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 8.555424210708822e-06,
+      "loss": 1.3732,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 1.0422,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.913,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.7626,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.7043,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.6356,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.6031,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5454,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4999,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5111,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8279639175257731,
+      "eval_loss": 0.4562760293483734,
+      "eval_runtime": 65.4867,
+      "eval_samples_per_second": 47.399,
+      "eval_steps_per_second": 11.85,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.511,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4849,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.5027,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.565,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4704,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4931,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4647,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4491,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4752,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3899,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8630798969072165,
+      "eval_loss": 0.3841710388660431,
+      "eval_runtime": 65.5356,
+      "eval_samples_per_second": 47.364,
+      "eval_steps_per_second": 11.841,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.477,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3964,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3763,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3928,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4235,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4024,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.4054,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3731,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3749,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3315,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8666237113402062,
+      "eval_loss": 0.3664605915546417,
+      "eval_runtime": 65.4939,
+      "eval_samples_per_second": 47.394,
+      "eval_steps_per_second": 11.848,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3572,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2822,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2383,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2685,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3076,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2832,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2976,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2826,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2585,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2816,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8753221649484536,
+      "eval_loss": 0.3534678518772125,
+      "eval_runtime": 65.5301,
+      "eval_samples_per_second": 47.368,
+      "eval_steps_per_second": 11.842,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2583,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2405,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2626,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1829,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.217,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2426,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2937,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.283,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2939,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2477,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8663015463917526,
+      "eval_loss": 0.37554699182510376,
+      "eval_runtime": 65.5483,
+      "eval_samples_per_second": 47.354,
+      "eval_steps_per_second": 11.839,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.3323,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2489,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2451,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2911,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2348,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2565,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1898,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2489,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2442,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2745,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8717783505154639,
+      "eval_loss": 0.3890085220336914,
+      "eval_runtime": 65.5564,
+      "eval_samples_per_second": 47.349,
+      "eval_steps_per_second": 11.837,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.208,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2804,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.186,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.206,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1658,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1576,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1845,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1601,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1397,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.109,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8814432989690721,
+      "eval_loss": 0.5400919318199158,
+      "eval_runtime": 65.5029,
+      "eval_samples_per_second": 47.387,
+      "eval_steps_per_second": 11.847,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1069,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1495,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1209,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.134,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1357,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1114,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1436,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1775,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1232,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.115,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8833762886597938,
+      "eval_loss": 0.5348294973373413,
+      "eval_runtime": 67.3607,
+      "eval_samples_per_second": 46.08,
+      "eval_steps_per_second": 11.52,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1429,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1612,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1565,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1188,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1576,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1832,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1643,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1668,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1984,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1839,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8424613402061856,
+      "eval_loss": 0.6889757513999939,
+      "eval_runtime": 65.603,
+      "eval_samples_per_second": 47.315,
+      "eval_steps_per_second": 11.829,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.2497,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1096,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.157,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0431,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0428,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1085,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0781,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1603,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0825,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0855,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.8759664948453608,
+      "eval_loss": 0.6595790982246399,
+      "eval_runtime": 65.5897,
+      "eval_samples_per_second": 47.324,
+      "eval_steps_per_second": 11.831,
+      "step": 2500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0614,
+      "step": 2525
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0686,
+      "step": 2550
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0716,
+      "step": 2575
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1359,
+      "step": 2600
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0863,
+      "step": 2625
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0915,
+      "step": 2650
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0595,
+      "step": 2675
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0714,
+      "step": 2700
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0914,
+      "step": 2725
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0774,
+      "step": 2750
+    },
+    {
+      "epoch": 3.54,
+      "eval_accuracy": 0.8740335051546392,
+      "eval_loss": 0.7505560517311096,
+      "eval_runtime": 65.5758,
+      "eval_samples_per_second": 47.335,
+      "eval_steps_per_second": 11.834,
+      "step": 2750
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0685,
+      "step": 2775
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.0943,
+      "step": 2800
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1181,
+      "step": 2825
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1302,
+      "step": 2850
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1058,
+      "step": 2875
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1235,
+      "step": 2900
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1278,
+      "step": 2925
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1172,
+      "step": 2950
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1461,
+      "step": 2975
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 1.7110848421417645e-05,
+      "loss": 0.1213,
+      "step": 3000
+    },
+    {
+      "epoch": 3.87,
+      "eval_accuracy": 0.8762886597938144,
+      "eval_loss": 0.6005998849868774,
+      "eval_runtime": 65.6002,
+      "eval_samples_per_second": 47.317,
+      "eval_steps_per_second": 11.829,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.262177729399808e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.7110848421417645e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-1/checkpoint-3000/training_args.bin b/run-1/checkpoint-3000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..109d901eeee674be35ec2b799f5af2df7bfd73e8
--- /dev/null
+++ b/run-1/checkpoint-3000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ff7ff5e497d2fd17b0b0a335d1e1e52fdf9ab64c61fc776fe6063c480e0a9f6
+size 4728
diff --git a/run-2/checkpoint-2500/config.json b/run-2/checkpoint-2500/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-2/checkpoint-2500/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-2/checkpoint-2500/model.safetensors b/run-2/checkpoint-2500/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..dd7c6fc1a7a1b8b4940782340e3d4369b15e2927
--- /dev/null
+++ b/run-2/checkpoint-2500/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0935e9b504369338f73d7b5f1be94c932055cbd56ef6d774f770dfbab5f141e9
+size 433279996
diff --git a/run-2/checkpoint-2500/optimizer.pt b/run-2/checkpoint-2500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8bf76e4f8701803151ce94632fd9796d78e9303a
--- /dev/null
+++ b/run-2/checkpoint-2500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a8f066b1a71ae52b3aa715094859018a19afc6f47f577e38281af9bc85a0c94
+size 866681082
diff --git a/run-2/checkpoint-2500/rng_state.pth b/run-2/checkpoint-2500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..678868d7ddbe48969a0ad74f8269b1e79c19b902
--- /dev/null
+++ b/run-2/checkpoint-2500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63399af7a409c183004eb47ca7dbc6f06cf71063012d7d165e74c89aaf482739
+size 14244
diff --git a/run-2/checkpoint-2500/scheduler.pt b/run-2/checkpoint-2500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b3d45eb1cb9bf5bb59bf9fda1e35a9cd229942aa
--- /dev/null
+++ b/run-2/checkpoint-2500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:390b240ea9153d81514422429b6a6dd0ee903146d58db9794a5b9cd5cbdd2201
+size 1064
diff --git a/run-2/checkpoint-2500/trainer_state.json b/run-2/checkpoint-2500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9e70243a20c20ba35daeae940b7fde6bb239fb16
--- /dev/null
+++ b/run-2/checkpoint-2500/trainer_state.json
@@ -0,0 +1,714 @@
+{
+  "best_metric": 0.8820876288659794,
+  "best_model_checkpoint": "bert_trainer/run-2/checkpoint-2000",
+  "epoch": 1.6118633139909737,
+  "eval_steps": 250,
+  "global_step": 2500,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.3571784276965265e-06,
+      "loss": 1.4073,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 1.1129,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.9656,
+      "step": 75
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.8417,
+      "step": 100
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7486,
+      "step": 125
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7111,
+      "step": 175
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7337,
+      "step": 200
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6518,
+      "step": 225
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6243,
+      "step": 250
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.8079896907216495,
+      "eval_loss": 0.5573504567146301,
+      "eval_runtime": 30.0553,
+      "eval_samples_per_second": 103.276,
+      "eval_steps_per_second": 25.819,
+      "step": 250
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5721,
+      "step": 275
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6388,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5117,
+      "step": 325
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6148,
+      "step": 350
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4858,
+      "step": 375
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.564,
+      "step": 400
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5233,
+      "step": 425
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4727,
+      "step": 450
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4875,
+      "step": 475
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4832,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8150773195876289,
+      "eval_loss": 0.4804709553718567,
+      "eval_runtime": 30.0129,
+      "eval_samples_per_second": 103.422,
+      "eval_steps_per_second": 25.856,
+      "step": 500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4925,
+      "step": 525
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5027,
+      "step": 550
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4571,
+      "step": 575
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4987,
+      "step": 600
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6134,
+      "step": 625
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5213,
+      "step": 650
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6241,
+      "step": 675
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4945,
+      "step": 700
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4353,
+      "step": 725
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4177,
+      "step": 750
+    },
+    {
+      "epoch": 0.48,
+      "eval_accuracy": 0.8411726804123711,
+      "eval_loss": 0.4728888273239136,
+      "eval_runtime": 30.0028,
+      "eval_samples_per_second": 103.457,
+      "eval_steps_per_second": 25.864,
+      "step": 750
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5676,
+      "step": 775
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3918,
+      "step": 800
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3912,
+      "step": 825
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4507,
+      "step": 850
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3633,
+      "step": 875
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4406,
+      "step": 900
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5013,
+      "step": 925
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4385,
+      "step": 950
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4561,
+      "step": 975
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3917,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8579252577319587,
+      "eval_loss": 0.39569276571273804,
+      "eval_runtime": 30.0411,
+      "eval_samples_per_second": 103.325,
+      "eval_steps_per_second": 25.831,
+      "step": 1000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4627,
+      "step": 1025
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4514,
+      "step": 1050
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4927,
+      "step": 1075
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3795,
+      "step": 1100
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3668,
+      "step": 1125
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4083,
+      "step": 1150
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4631,
+      "step": 1175
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4035,
+      "step": 1200
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3476,
+      "step": 1225
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5089,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "eval_accuracy": 0.8479381443298969,
+      "eval_loss": 0.4126531481742859,
+      "eval_runtime": 30.0563,
+      "eval_samples_per_second": 103.273,
+      "eval_steps_per_second": 25.818,
+      "step": 1250
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4805,
+      "step": 1275
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3949,
+      "step": 1300
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3993,
+      "step": 1325
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4719,
+      "step": 1350
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4545,
+      "step": 1375
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3411,
+      "step": 1400
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4989,
+      "step": 1425
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4016,
+      "step": 1450
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3826,
+      "step": 1475
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.376,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8724226804123711,
+      "eval_loss": 0.3830737769603729,
+      "eval_runtime": 30.0147,
+      "eval_samples_per_second": 103.416,
+      "eval_steps_per_second": 25.854,
+      "step": 1500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4116,
+      "step": 1525
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3377,
+      "step": 1550
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2881,
+      "step": 1575
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2914,
+      "step": 1600
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2836,
+      "step": 1625
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3338,
+      "step": 1650
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3104,
+      "step": 1675
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2209,
+      "step": 1700
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3869,
+      "step": 1725
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2869,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.8756443298969072,
+      "eval_loss": 0.49572044610977173,
+      "eval_runtime": 30.0097,
+      "eval_samples_per_second": 103.433,
+      "eval_steps_per_second": 25.858,
+      "step": 1750
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2977,
+      "step": 1775
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3517,
+      "step": 1825
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.342,
+      "step": 1850
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.248,
+      "step": 1875
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3462,
+      "step": 1900
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2848,
+      "step": 1925
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3189,
+      "step": 1950
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2977,
+      "step": 1975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2707,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8820876288659794,
+      "eval_loss": 0.4196120500564575,
+      "eval_runtime": 30.0474,
+      "eval_samples_per_second": 103.303,
+      "eval_steps_per_second": 25.826,
+      "step": 2000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3122,
+      "step": 2025
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2782,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2473,
+      "step": 2075
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3009,
+      "step": 2100
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2717,
+      "step": 2125
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4527,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1456,
+      "step": 2175
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3033,
+      "step": 2200
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2462,
+      "step": 2225
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1881,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "eval_accuracy": 0.876610824742268,
+      "eval_loss": 0.48145169019699097,
+      "eval_runtime": 30.011,
+      "eval_samples_per_second": 103.429,
+      "eval_steps_per_second": 25.857,
+      "step": 2250
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3002,
+      "step": 2275
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2344,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4302,
+      "step": 2325
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1932,
+      "step": 2350
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2573,
+      "step": 2375
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3697,
+      "step": 2400
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3531,
+      "step": 2425
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3102,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3344,
+      "step": 2475
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2301,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8646907216494846,
+      "eval_loss": 0.5154445171356201,
+      "eval_runtime": 30.0652,
+      "eval_samples_per_second": 103.242,
+      "eval_steps_per_second": 25.811,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 250,
+  "total_flos": 5261836612995072.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.4714356855393053e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 8
+  }
+}
diff --git a/run-2/checkpoint-2500/training_args.bin b/run-2/checkpoint-2500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..e230b2b0691c2ba955d08c27b01c43281c9d4fbc
--- /dev/null
+++ b/run-2/checkpoint-2500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40bef76d9ba6b7f9c82fb91b5dea666f234a2847e28cb2faf44f5ca8e9ea5d19
+size 4728
diff --git a/run-2/checkpoint-2750/config.json b/run-2/checkpoint-2750/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-2/checkpoint-2750/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-2/checkpoint-2750/model.safetensors b/run-2/checkpoint-2750/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0c32428bafca9e21ad0395f6a18bd0aa9bce112e
--- /dev/null
+++ b/run-2/checkpoint-2750/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c210fb994a6336c42ca80164f32a4333330de92d54271f83d4b0eddea8243960
+size 433279996
diff --git a/run-2/checkpoint-2750/optimizer.pt b/run-2/checkpoint-2750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..37023a45550ff1d4f08e47282be949861db1ce4b
--- /dev/null
+++ b/run-2/checkpoint-2750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18db280778e9705d46ece0c5745141d160acca15de8c7f69593f0abeb9bc6049
+size 866681082
diff --git a/run-2/checkpoint-2750/rng_state.pth b/run-2/checkpoint-2750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ecabdb8c07de6e55c7332facacb47f6edf2a41ca
--- /dev/null
+++ b/run-2/checkpoint-2750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd2745a1e1a49ccd51b48496c3e2a5d498e0bdad8fb3f45b3589cf70267b4702
+size 14244
diff --git a/run-2/checkpoint-2750/scheduler.pt b/run-2/checkpoint-2750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5dc9e10f82f2385da43c97d0d649cda49d18620e
--- /dev/null
+++ b/run-2/checkpoint-2750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2649d982abf02498484134a6453860cae843bf59a4a4be51c1d55980e8ec4c87
+size 1064
diff --git a/run-2/checkpoint-2750/trainer_state.json b/run-2/checkpoint-2750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..438049d0b4c5bb942f3efe4db70cbc141b4e1387
--- /dev/null
+++ b/run-2/checkpoint-2750/trainer_state.json
@@ -0,0 +1,783 @@
+{
+  "best_metric": 0.8820876288659794,
+  "best_model_checkpoint": "bert_trainer/run-2/checkpoint-2000",
+  "epoch": 1.773049645390071,
+  "eval_steps": 250,
+  "global_step": 2750,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.3571784276965265e-06,
+      "loss": 1.4073,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 1.1129,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.9656,
+      "step": 75
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.8417,
+      "step": 100
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7486,
+      "step": 125
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7111,
+      "step": 175
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7337,
+      "step": 200
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6518,
+      "step": 225
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6243,
+      "step": 250
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.8079896907216495,
+      "eval_loss": 0.5573504567146301,
+      "eval_runtime": 30.0553,
+      "eval_samples_per_second": 103.276,
+      "eval_steps_per_second": 25.819,
+      "step": 250
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5721,
+      "step": 275
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6388,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5117,
+      "step": 325
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6148,
+      "step": 350
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4858,
+      "step": 375
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.564,
+      "step": 400
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5233,
+      "step": 425
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4727,
+      "step": 450
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4875,
+      "step": 475
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4832,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8150773195876289,
+      "eval_loss": 0.4804709553718567,
+      "eval_runtime": 30.0129,
+      "eval_samples_per_second": 103.422,
+      "eval_steps_per_second": 25.856,
+      "step": 500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4925,
+      "step": 525
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5027,
+      "step": 550
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4571,
+      "step": 575
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4987,
+      "step": 600
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6134,
+      "step": 625
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5213,
+      "step": 650
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6241,
+      "step": 675
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4945,
+      "step": 700
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4353,
+      "step": 725
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4177,
+      "step": 750
+    },
+    {
+      "epoch": 0.48,
+      "eval_accuracy": 0.8411726804123711,
+      "eval_loss": 0.4728888273239136,
+      "eval_runtime": 30.0028,
+      "eval_samples_per_second": 103.457,
+      "eval_steps_per_second": 25.864,
+      "step": 750
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5676,
+      "step": 775
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3918,
+      "step": 800
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3912,
+      "step": 825
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4507,
+      "step": 850
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3633,
+      "step": 875
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4406,
+      "step": 900
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5013,
+      "step": 925
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4385,
+      "step": 950
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4561,
+      "step": 975
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3917,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8579252577319587,
+      "eval_loss": 0.39569276571273804,
+      "eval_runtime": 30.0411,
+      "eval_samples_per_second": 103.325,
+      "eval_steps_per_second": 25.831,
+      "step": 1000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4627,
+      "step": 1025
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4514,
+      "step": 1050
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4927,
+      "step": 1075
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3795,
+      "step": 1100
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3668,
+      "step": 1125
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4083,
+      "step": 1150
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4631,
+      "step": 1175
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4035,
+      "step": 1200
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3476,
+      "step": 1225
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5089,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "eval_accuracy": 0.8479381443298969,
+      "eval_loss": 0.4126531481742859,
+      "eval_runtime": 30.0563,
+      "eval_samples_per_second": 103.273,
+      "eval_steps_per_second": 25.818,
+      "step": 1250
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4805,
+      "step": 1275
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3949,
+      "step": 1300
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3993,
+      "step": 1325
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4719,
+      "step": 1350
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4545,
+      "step": 1375
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3411,
+      "step": 1400
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4989,
+      "step": 1425
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4016,
+      "step": 1450
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3826,
+      "step": 1475
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.376,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8724226804123711,
+      "eval_loss": 0.3830737769603729,
+      "eval_runtime": 30.0147,
+      "eval_samples_per_second": 103.416,
+      "eval_steps_per_second": 25.854,
+      "step": 1500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4116,
+      "step": 1525
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3377,
+      "step": 1550
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2881,
+      "step": 1575
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2914,
+      "step": 1600
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2836,
+      "step": 1625
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3338,
+      "step": 1650
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3104,
+      "step": 1675
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2209,
+      "step": 1700
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3869,
+      "step": 1725
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2869,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.8756443298969072,
+      "eval_loss": 0.49572044610977173,
+      "eval_runtime": 30.0097,
+      "eval_samples_per_second": 103.433,
+      "eval_steps_per_second": 25.858,
+      "step": 1750
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2977,
+      "step": 1775
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3517,
+      "step": 1825
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.342,
+      "step": 1850
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.248,
+      "step": 1875
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3462,
+      "step": 1900
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2848,
+      "step": 1925
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3189,
+      "step": 1950
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2977,
+      "step": 1975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2707,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8820876288659794,
+      "eval_loss": 0.4196120500564575,
+      "eval_runtime": 30.0474,
+      "eval_samples_per_second": 103.303,
+      "eval_steps_per_second": 25.826,
+      "step": 2000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3122,
+      "step": 2025
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2782,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2473,
+      "step": 2075
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3009,
+      "step": 2100
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2717,
+      "step": 2125
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4527,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1456,
+      "step": 2175
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3033,
+      "step": 2200
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2462,
+      "step": 2225
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1881,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "eval_accuracy": 0.876610824742268,
+      "eval_loss": 0.48145169019699097,
+      "eval_runtime": 30.011,
+      "eval_samples_per_second": 103.429,
+      "eval_steps_per_second": 25.857,
+      "step": 2250
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3002,
+      "step": 2275
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2344,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4302,
+      "step": 2325
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1932,
+      "step": 2350
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2573,
+      "step": 2375
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3697,
+      "step": 2400
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3531,
+      "step": 2425
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3102,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3344,
+      "step": 2475
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2301,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8646907216494846,
+      "eval_loss": 0.5154445171356201,
+      "eval_runtime": 30.0652,
+      "eval_samples_per_second": 103.242,
+      "eval_steps_per_second": 25.811,
+      "step": 2500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3972,
+      "step": 2525
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3339,
+      "step": 2550
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.334,
+      "step": 2575
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2085,
+      "step": 2600
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2737,
+      "step": 2625
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3618,
+      "step": 2650
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3648,
+      "step": 2675
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4173,
+      "step": 2700
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2908,
+      "step": 2725
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2368,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "eval_accuracy": 0.8817654639175257,
+      "eval_loss": 0.4021648168563843,
+      "eval_runtime": 30.0057,
+      "eval_samples_per_second": 103.447,
+      "eval_steps_per_second": 25.862,
+      "step": 2750
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 250,
+  "total_flos": 5788072897923072.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.4714356855393053e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 8
+  }
+}
diff --git a/run-2/checkpoint-2750/training_args.bin b/run-2/checkpoint-2750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..e230b2b0691c2ba955d08c27b01c43281c9d4fbc
--- /dev/null
+++ b/run-2/checkpoint-2750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40bef76d9ba6b7f9c82fb91b5dea666f234a2847e28cb2faf44f5ca8e9ea5d19
+size 4728
diff --git a/run-2/checkpoint-3000/config.json b/run-2/checkpoint-3000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-2/checkpoint-3000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-2/checkpoint-3000/model.safetensors b/run-2/checkpoint-3000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..987fd84b572e548945a6a5d9dcaf8b3a213c9c51
--- /dev/null
+++ b/run-2/checkpoint-3000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1324cd5b343ad24109e6853a17728a3c5d9c7424de3134f62b0784f1f159efc0
+size 433279996
diff --git a/run-2/checkpoint-3000/optimizer.pt b/run-2/checkpoint-3000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e5b0448c1e0287fc702490f2d9ab6ace8ee7a655
--- /dev/null
+++ b/run-2/checkpoint-3000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fc5a0a580536a36fe6b033505639d32cd0293f8d41ff8aca99ad73d7a97adfb
+size 866681082
diff --git a/run-2/checkpoint-3000/rng_state.pth b/run-2/checkpoint-3000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d717cb03d14eef32afe17820643b27cad107af13
--- /dev/null
+++ b/run-2/checkpoint-3000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a039281de77e6b9a31e662411886dca5f1d510240f4f6e504c9f52ef9b31a88b
+size 14244
diff --git a/run-2/checkpoint-3000/scheduler.pt b/run-2/checkpoint-3000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8c2c9740c2d9a0d3e37317dac79d7846c0ec9e50
--- /dev/null
+++ b/run-2/checkpoint-3000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7938821ec967969ef940b4f317cf1e8213f672949806cb73a03d1ac52e895231
+size 1064
diff --git a/run-2/checkpoint-3000/trainer_state.json b/run-2/checkpoint-3000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..1d61c4930ff0f9bf2842f96845dd688c92526f46
--- /dev/null
+++ b/run-2/checkpoint-3000/trainer_state.json
@@ -0,0 +1,852 @@
+{
+  "best_metric": 0.882409793814433,
+  "best_model_checkpoint": "bert_trainer/run-2/checkpoint-3000",
+  "epoch": 1.9342359767891684,
+  "eval_steps": 250,
+  "global_step": 3000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.3571784276965265e-06,
+      "loss": 1.4073,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 1.1129,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.9656,
+      "step": 75
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.8417,
+      "step": 100
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7486,
+      "step": 125
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.8348,
+      "step": 150
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7111,
+      "step": 175
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.7337,
+      "step": 200
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6518,
+      "step": 225
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6243,
+      "step": 250
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.8079896907216495,
+      "eval_loss": 0.5573504567146301,
+      "eval_runtime": 30.0553,
+      "eval_samples_per_second": 103.276,
+      "eval_steps_per_second": 25.819,
+      "step": 250
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5721,
+      "step": 275
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6388,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5117,
+      "step": 325
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6148,
+      "step": 350
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4858,
+      "step": 375
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.564,
+      "step": 400
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5233,
+      "step": 425
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4727,
+      "step": 450
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4875,
+      "step": 475
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4832,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8150773195876289,
+      "eval_loss": 0.4804709553718567,
+      "eval_runtime": 30.0129,
+      "eval_samples_per_second": 103.422,
+      "eval_steps_per_second": 25.856,
+      "step": 500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4925,
+      "step": 525
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5027,
+      "step": 550
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4571,
+      "step": 575
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4987,
+      "step": 600
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6134,
+      "step": 625
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5213,
+      "step": 650
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.6241,
+      "step": 675
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4945,
+      "step": 700
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4353,
+      "step": 725
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4177,
+      "step": 750
+    },
+    {
+      "epoch": 0.48,
+      "eval_accuracy": 0.8411726804123711,
+      "eval_loss": 0.4728888273239136,
+      "eval_runtime": 30.0028,
+      "eval_samples_per_second": 103.457,
+      "eval_steps_per_second": 25.864,
+      "step": 750
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5676,
+      "step": 775
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3918,
+      "step": 800
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3912,
+      "step": 825
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4507,
+      "step": 850
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3633,
+      "step": 875
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4406,
+      "step": 900
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5013,
+      "step": 925
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4385,
+      "step": 950
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4561,
+      "step": 975
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3917,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8579252577319587,
+      "eval_loss": 0.39569276571273804,
+      "eval_runtime": 30.0411,
+      "eval_samples_per_second": 103.325,
+      "eval_steps_per_second": 25.831,
+      "step": 1000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4627,
+      "step": 1025
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4514,
+      "step": 1050
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4927,
+      "step": 1075
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3795,
+      "step": 1100
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3668,
+      "step": 1125
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4083,
+      "step": 1150
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4631,
+      "step": 1175
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4035,
+      "step": 1200
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3476,
+      "step": 1225
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.5089,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "eval_accuracy": 0.8479381443298969,
+      "eval_loss": 0.4126531481742859,
+      "eval_runtime": 30.0563,
+      "eval_samples_per_second": 103.273,
+      "eval_steps_per_second": 25.818,
+      "step": 1250
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4805,
+      "step": 1275
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3949,
+      "step": 1300
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3993,
+      "step": 1325
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4719,
+      "step": 1350
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4545,
+      "step": 1375
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3411,
+      "step": 1400
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4989,
+      "step": 1425
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4016,
+      "step": 1450
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3826,
+      "step": 1475
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.376,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8724226804123711,
+      "eval_loss": 0.3830737769603729,
+      "eval_runtime": 30.0147,
+      "eval_samples_per_second": 103.416,
+      "eval_steps_per_second": 25.854,
+      "step": 1500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4116,
+      "step": 1525
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3377,
+      "step": 1550
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2881,
+      "step": 1575
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2914,
+      "step": 1600
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2836,
+      "step": 1625
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3338,
+      "step": 1650
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3104,
+      "step": 1675
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2209,
+      "step": 1700
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3869,
+      "step": 1725
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2869,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.8756443298969072,
+      "eval_loss": 0.49572044610977173,
+      "eval_runtime": 30.0097,
+      "eval_samples_per_second": 103.433,
+      "eval_steps_per_second": 25.858,
+      "step": 1750
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2977,
+      "step": 1775
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3765,
+      "step": 1800
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3517,
+      "step": 1825
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.342,
+      "step": 1850
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.248,
+      "step": 1875
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3462,
+      "step": 1900
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2848,
+      "step": 1925
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3189,
+      "step": 1950
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2977,
+      "step": 1975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2707,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8820876288659794,
+      "eval_loss": 0.4196120500564575,
+      "eval_runtime": 30.0474,
+      "eval_samples_per_second": 103.303,
+      "eval_steps_per_second": 25.826,
+      "step": 2000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3122,
+      "step": 2025
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2782,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2473,
+      "step": 2075
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3009,
+      "step": 2100
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2717,
+      "step": 2125
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4527,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1456,
+      "step": 2175
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3033,
+      "step": 2200
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2462,
+      "step": 2225
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1881,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "eval_accuracy": 0.876610824742268,
+      "eval_loss": 0.48145169019699097,
+      "eval_runtime": 30.011,
+      "eval_samples_per_second": 103.429,
+      "eval_steps_per_second": 25.857,
+      "step": 2250
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3002,
+      "step": 2275
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2344,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4302,
+      "step": 2325
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1932,
+      "step": 2350
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2573,
+      "step": 2375
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3697,
+      "step": 2400
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3531,
+      "step": 2425
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3102,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3344,
+      "step": 2475
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2301,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8646907216494846,
+      "eval_loss": 0.5154445171356201,
+      "eval_runtime": 30.0652,
+      "eval_samples_per_second": 103.242,
+      "eval_steps_per_second": 25.811,
+      "step": 2500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3972,
+      "step": 2525
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3339,
+      "step": 2550
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.334,
+      "step": 2575
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2085,
+      "step": 2600
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2737,
+      "step": 2625
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3618,
+      "step": 2650
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3648,
+      "step": 2675
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.4173,
+      "step": 2700
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2908,
+      "step": 2725
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2368,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "eval_accuracy": 0.8817654639175257,
+      "eval_loss": 0.4021648168563843,
+      "eval_runtime": 30.0057,
+      "eval_samples_per_second": 103.447,
+      "eval_steps_per_second": 25.862,
+      "step": 2750
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2185,
+      "step": 2775
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3783,
+      "step": 2800
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2199,
+      "step": 2825
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2365,
+      "step": 2850
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2724,
+      "step": 2875
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2973,
+      "step": 2900
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.1881,
+      "step": 2925
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2876,
+      "step": 2950
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.2775,
+      "step": 2975
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.4714356855393053e-05,
+      "loss": 0.3342,
+      "step": 3000
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.882409793814433,
+      "eval_loss": 0.46482139825820923,
+      "eval_runtime": 30.0087,
+      "eval_samples_per_second": 103.437,
+      "eval_steps_per_second": 25.859,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 250,
+  "total_flos": 6314309182851072.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.4714356855393053e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 8
+  }
+}
diff --git a/run-2/checkpoint-3000/training_args.bin b/run-2/checkpoint-3000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..e230b2b0691c2ba955d08c27b01c43281c9d4fbc
--- /dev/null
+++ b/run-2/checkpoint-3000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40bef76d9ba6b7f9c82fb91b5dea666f234a2847e28cb2faf44f5ca8e9ea5d19
+size 4728
diff --git a/run-3/checkpoint-2500/config.json b/run-3/checkpoint-2500/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-3/checkpoint-2500/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-3/checkpoint-2500/model.safetensors b/run-3/checkpoint-2500/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d37b0bc616953f5a9dcd49129e8e6195c50af76b
--- /dev/null
+++ b/run-3/checkpoint-2500/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0cc7504acb0df38cf2b8397879c678cc053c99d057855e90ceacb0fc02033b9
+size 433279996
diff --git a/run-3/checkpoint-2500/optimizer.pt b/run-3/checkpoint-2500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..799acdddf0314a85dc12a777caa4e600b65fee8b
--- /dev/null
+++ b/run-3/checkpoint-2500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8aefff0ed89024633b56e88d5acb54c3869bf92689d5a6a722d49dbf6579f91a
+size 866681082
diff --git a/run-3/checkpoint-2500/rng_state.pth b/run-3/checkpoint-2500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..678868d7ddbe48969a0ad74f8269b1e79c19b902
--- /dev/null
+++ b/run-3/checkpoint-2500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63399af7a409c183004eb47ca7dbc6f06cf71063012d7d165e74c89aaf482739
+size 14244
diff --git a/run-3/checkpoint-2500/scheduler.pt b/run-3/checkpoint-2500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..57daf78c38c291afbf56264150d165a16b1b1585
--- /dev/null
+++ b/run-3/checkpoint-2500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64d403cc14cfc7d49ef4f76591323ffab474e41871c823468df2cfb7721d0f76
+size 1064
diff --git a/run-3/checkpoint-2500/trainer_state.json b/run-3/checkpoint-2500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..e1a547730be2031e24914b76bf9f811d0adfbe5d
--- /dev/null
+++ b/run-3/checkpoint-2500/trainer_state.json
@@ -0,0 +1,714 @@
+{
+  "best_metric": 0.8730670103092784,
+  "best_model_checkpoint": "bert_trainer/run-3/checkpoint-2000",
+  "epoch": 1.6118633139909737,
+  "eval_steps": 250,
+  "global_step": 2500,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3880483293934951e-05,
+      "loss": 1.3534,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 1.0289,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.8774,
+      "step": 75
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.7025,
+      "step": 100
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.8032,
+      "step": 125
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.808,
+      "step": 150
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.651,
+      "step": 175
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.6946,
+      "step": 200
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.6587,
+      "step": 225
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5824,
+      "step": 250
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.8112113402061856,
+      "eval_loss": 0.5165922045707703,
+      "eval_runtime": 30.3844,
+      "eval_samples_per_second": 102.158,
+      "eval_steps_per_second": 6.385,
+      "step": 250
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.546,
+      "step": 275
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5713,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4633,
+      "step": 325
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5839,
+      "step": 350
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4685,
+      "step": 375
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5632,
+      "step": 400
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.52,
+      "step": 425
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5565,
+      "step": 450
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5394,
+      "step": 475
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4944,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8244201030927835,
+      "eval_loss": 0.4521254003047943,
+      "eval_runtime": 30.375,
+      "eval_samples_per_second": 102.189,
+      "eval_steps_per_second": 6.387,
+      "step": 500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5233,
+      "step": 525
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.475,
+      "step": 550
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4169,
+      "step": 575
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4961,
+      "step": 600
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5372,
+      "step": 625
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5151,
+      "step": 650
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5491,
+      "step": 675
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4596,
+      "step": 700
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4722,
+      "step": 725
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3632,
+      "step": 750
+    },
+    {
+      "epoch": 0.48,
+      "eval_accuracy": 0.8344072164948454,
+      "eval_loss": 0.48342078924179077,
+      "eval_runtime": 30.4129,
+      "eval_samples_per_second": 102.062,
+      "eval_steps_per_second": 6.379,
+      "step": 750
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5665,
+      "step": 775
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3748,
+      "step": 800
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4397,
+      "step": 825
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4141,
+      "step": 850
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4029,
+      "step": 875
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4191,
+      "step": 900
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5094,
+      "step": 925
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4308,
+      "step": 950
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3921,
+      "step": 975
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4381,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8588917525773195,
+      "eval_loss": 0.4446374177932739,
+      "eval_runtime": 29.9606,
+      "eval_samples_per_second": 103.603,
+      "eval_steps_per_second": 6.475,
+      "step": 1000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4766,
+      "step": 1025
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4957,
+      "step": 1050
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4285,
+      "step": 1075
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3494,
+      "step": 1100
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4066,
+      "step": 1125
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3833,
+      "step": 1150
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4553,
+      "step": 1175
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4185,
+      "step": 1200
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3519,
+      "step": 1225
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4848,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "eval_accuracy": 0.8582474226804123,
+      "eval_loss": 0.408128023147583,
+      "eval_runtime": 29.8798,
+      "eval_samples_per_second": 103.883,
+      "eval_steps_per_second": 6.493,
+      "step": 1250
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.539,
+      "step": 1275
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3632,
+      "step": 1300
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3861,
+      "step": 1325
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5181,
+      "step": 1350
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4158,
+      "step": 1375
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3584,
+      "step": 1400
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4162,
+      "step": 1425
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4607,
+      "step": 1450
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4146,
+      "step": 1475
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3954,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8714561855670103,
+      "eval_loss": 0.38059666752815247,
+      "eval_runtime": 29.8772,
+      "eval_samples_per_second": 103.892,
+      "eval_steps_per_second": 6.493,
+      "step": 1500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3881,
+      "step": 1525
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3825,
+      "step": 1550
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3082,
+      "step": 1575
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3367,
+      "step": 1600
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3191,
+      "step": 1625
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3298,
+      "step": 1650
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2631,
+      "step": 1675
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1897,
+      "step": 1700
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3901,
+      "step": 1725
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3268,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.8714561855670103,
+      "eval_loss": 0.5426503419876099,
+      "eval_runtime": 29.922,
+      "eval_samples_per_second": 103.737,
+      "eval_steps_per_second": 6.484,
+      "step": 1750
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2771,
+      "step": 1775
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3346,
+      "step": 1800
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.312,
+      "step": 1825
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3482,
+      "step": 1850
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3186,
+      "step": 1875
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.317,
+      "step": 1900
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3214,
+      "step": 1925
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2073,
+      "step": 1950
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3709,
+      "step": 1975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2492,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8730670103092784,
+      "eval_loss": 0.4012953042984009,
+      "eval_runtime": 29.8764,
+      "eval_samples_per_second": 103.895,
+      "eval_steps_per_second": 6.493,
+      "step": 2000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2831,
+      "step": 2025
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2479,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2467,
+      "step": 2075
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.332,
+      "step": 2100
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2565,
+      "step": 2125
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4016,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1485,
+      "step": 2175
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3333,
+      "step": 2200
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3541,
+      "step": 2225
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1829,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "eval_accuracy": 0.8704896907216495,
+      "eval_loss": 0.49169790744781494,
+      "eval_runtime": 29.9217,
+      "eval_samples_per_second": 103.737,
+      "eval_steps_per_second": 6.484,
+      "step": 2250
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2888,
+      "step": 2275
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2472,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3547,
+      "step": 2325
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1869,
+      "step": 2350
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3094,
+      "step": 2375
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.314,
+      "step": 2400
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3161,
+      "step": 2425
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3445,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4138,
+      "step": 2475
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2214,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.852770618556701,
+      "eval_loss": 0.5358509421348572,
+      "eval_runtime": 29.9717,
+      "eval_samples_per_second": 103.564,
+      "eval_steps_per_second": 6.473,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 250,
+  "total_flos": 5261836612995072.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 2.7760966587869903e-05,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 8
+  }
+}
diff --git a/run-3/checkpoint-2500/training_args.bin b/run-3/checkpoint-2500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..50cc4ec23d371b30ae6512db04b9c655f9dcecd3
--- /dev/null
+++ b/run-3/checkpoint-2500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2175af5e4c44e0db4ee2b9299c68887232a3ab86bd73b88fdec63aaf1669da81
+size 4728
diff --git a/run-3/checkpoint-2750/config.json b/run-3/checkpoint-2750/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-3/checkpoint-2750/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-3/checkpoint-2750/model.safetensors b/run-3/checkpoint-2750/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a756a2e7d8572f1c45af4d33d96c71911ca547ed
--- /dev/null
+++ b/run-3/checkpoint-2750/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:677860fa8e1e6f97f106d175e4bf43a3ee5a373c005215685688306dcfa72a49
+size 433279996
diff --git a/run-3/checkpoint-2750/optimizer.pt b/run-3/checkpoint-2750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5c6366ca682433d334b91ae4ddf51642e26b44b7
--- /dev/null
+++ b/run-3/checkpoint-2750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35f9e541a8c8ecfe7e4c311480c4f8686b0203294e2325962ab05d3786317e2d
+size 866681082
diff --git a/run-3/checkpoint-2750/rng_state.pth b/run-3/checkpoint-2750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ecabdb8c07de6e55c7332facacb47f6edf2a41ca
--- /dev/null
+++ b/run-3/checkpoint-2750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd2745a1e1a49ccd51b48496c3e2a5d498e0bdad8fb3f45b3589cf70267b4702
+size 14244
diff --git a/run-3/checkpoint-2750/scheduler.pt b/run-3/checkpoint-2750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..db757bdc64cc1181d0dc1fa9756cbb6d8d71361d
--- /dev/null
+++ b/run-3/checkpoint-2750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9d874b36e5555754e89b613b4b82c5013e5f20218f93ef6d523732fbbfbbbb3
+size 1064
diff --git a/run-3/checkpoint-2750/trainer_state.json b/run-3/checkpoint-2750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec6a3a82fe542f5dcf423ffeffd2e5a2deee9fc7
--- /dev/null
+++ b/run-3/checkpoint-2750/trainer_state.json
@@ -0,0 +1,783 @@
+{
+  "best_metric": 0.884020618556701,
+  "best_model_checkpoint": "bert_trainer/run-3/checkpoint-2750",
+  "epoch": 1.773049645390071,
+  "eval_steps": 250,
+  "global_step": 2750,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3880483293934951e-05,
+      "loss": 1.3534,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 1.0289,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.8774,
+      "step": 75
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.7025,
+      "step": 100
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.8032,
+      "step": 125
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.808,
+      "step": 150
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.651,
+      "step": 175
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.6946,
+      "step": 200
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.6587,
+      "step": 225
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5824,
+      "step": 250
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.8112113402061856,
+      "eval_loss": 0.5165922045707703,
+      "eval_runtime": 30.3844,
+      "eval_samples_per_second": 102.158,
+      "eval_steps_per_second": 6.385,
+      "step": 250
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.546,
+      "step": 275
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5713,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4633,
+      "step": 325
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5839,
+      "step": 350
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4685,
+      "step": 375
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5632,
+      "step": 400
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.52,
+      "step": 425
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5565,
+      "step": 450
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5394,
+      "step": 475
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4944,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8244201030927835,
+      "eval_loss": 0.4521254003047943,
+      "eval_runtime": 30.375,
+      "eval_samples_per_second": 102.189,
+      "eval_steps_per_second": 6.387,
+      "step": 500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5233,
+      "step": 525
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.475,
+      "step": 550
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4169,
+      "step": 575
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4961,
+      "step": 600
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5372,
+      "step": 625
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5151,
+      "step": 650
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5491,
+      "step": 675
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4596,
+      "step": 700
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4722,
+      "step": 725
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3632,
+      "step": 750
+    },
+    {
+      "epoch": 0.48,
+      "eval_accuracy": 0.8344072164948454,
+      "eval_loss": 0.48342078924179077,
+      "eval_runtime": 30.4129,
+      "eval_samples_per_second": 102.062,
+      "eval_steps_per_second": 6.379,
+      "step": 750
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5665,
+      "step": 775
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3748,
+      "step": 800
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4397,
+      "step": 825
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4141,
+      "step": 850
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4029,
+      "step": 875
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4191,
+      "step": 900
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5094,
+      "step": 925
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4308,
+      "step": 950
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3921,
+      "step": 975
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4381,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8588917525773195,
+      "eval_loss": 0.4446374177932739,
+      "eval_runtime": 29.9606,
+      "eval_samples_per_second": 103.603,
+      "eval_steps_per_second": 6.475,
+      "step": 1000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4766,
+      "step": 1025
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4957,
+      "step": 1050
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4285,
+      "step": 1075
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3494,
+      "step": 1100
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4066,
+      "step": 1125
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3833,
+      "step": 1150
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4553,
+      "step": 1175
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4185,
+      "step": 1200
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3519,
+      "step": 1225
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4848,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "eval_accuracy": 0.8582474226804123,
+      "eval_loss": 0.408128023147583,
+      "eval_runtime": 29.8798,
+      "eval_samples_per_second": 103.883,
+      "eval_steps_per_second": 6.493,
+      "step": 1250
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.539,
+      "step": 1275
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3632,
+      "step": 1300
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3861,
+      "step": 1325
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5181,
+      "step": 1350
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4158,
+      "step": 1375
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3584,
+      "step": 1400
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4162,
+      "step": 1425
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4607,
+      "step": 1450
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4146,
+      "step": 1475
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3954,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8714561855670103,
+      "eval_loss": 0.38059666752815247,
+      "eval_runtime": 29.8772,
+      "eval_samples_per_second": 103.892,
+      "eval_steps_per_second": 6.493,
+      "step": 1500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3881,
+      "step": 1525
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3825,
+      "step": 1550
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3082,
+      "step": 1575
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3367,
+      "step": 1600
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3191,
+      "step": 1625
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3298,
+      "step": 1650
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2631,
+      "step": 1675
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1897,
+      "step": 1700
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3901,
+      "step": 1725
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3268,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.8714561855670103,
+      "eval_loss": 0.5426503419876099,
+      "eval_runtime": 29.922,
+      "eval_samples_per_second": 103.737,
+      "eval_steps_per_second": 6.484,
+      "step": 1750
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2771,
+      "step": 1775
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3346,
+      "step": 1800
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.312,
+      "step": 1825
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3482,
+      "step": 1850
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3186,
+      "step": 1875
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.317,
+      "step": 1900
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3214,
+      "step": 1925
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2073,
+      "step": 1950
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3709,
+      "step": 1975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2492,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8730670103092784,
+      "eval_loss": 0.4012953042984009,
+      "eval_runtime": 29.8764,
+      "eval_samples_per_second": 103.895,
+      "eval_steps_per_second": 6.493,
+      "step": 2000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2831,
+      "step": 2025
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2479,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2467,
+      "step": 2075
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.332,
+      "step": 2100
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2565,
+      "step": 2125
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4016,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1485,
+      "step": 2175
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3333,
+      "step": 2200
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3541,
+      "step": 2225
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1829,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "eval_accuracy": 0.8704896907216495,
+      "eval_loss": 0.49169790744781494,
+      "eval_runtime": 29.9217,
+      "eval_samples_per_second": 103.737,
+      "eval_steps_per_second": 6.484,
+      "step": 2250
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2888,
+      "step": 2275
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2472,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3547,
+      "step": 2325
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1869,
+      "step": 2350
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3094,
+      "step": 2375
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.314,
+      "step": 2400
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3161,
+      "step": 2425
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3445,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4138,
+      "step": 2475
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2214,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.852770618556701,
+      "eval_loss": 0.5358509421348572,
+      "eval_runtime": 29.9717,
+      "eval_samples_per_second": 103.564,
+      "eval_steps_per_second": 6.473,
+      "step": 2500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3897,
+      "step": 2525
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3371,
+      "step": 2550
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3278,
+      "step": 2575
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2226,
+      "step": 2600
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2836,
+      "step": 2625
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3236,
+      "step": 2650
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3125,
+      "step": 2675
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3267,
+      "step": 2700
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2427,
+      "step": 2725
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2973,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "eval_accuracy": 0.884020618556701,
+      "eval_loss": 0.42780932784080505,
+      "eval_runtime": 29.8873,
+      "eval_samples_per_second": 103.857,
+      "eval_steps_per_second": 6.491,
+      "step": 2750
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 250,
+  "total_flos": 5788072897923072.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 2.7760966587869903e-05,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 8
+  }
+}
diff --git a/run-3/checkpoint-2750/training_args.bin b/run-3/checkpoint-2750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..50cc4ec23d371b30ae6512db04b9c655f9dcecd3
--- /dev/null
+++ b/run-3/checkpoint-2750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2175af5e4c44e0db4ee2b9299c68887232a3ab86bd73b88fdec63aaf1669da81
+size 4728
diff --git a/run-3/checkpoint-3000/config.json b/run-3/checkpoint-3000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-3/checkpoint-3000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-3/checkpoint-3000/model.safetensors b/run-3/checkpoint-3000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ba42277f4f96cd3879fa9109321eaeed83957229
--- /dev/null
+++ b/run-3/checkpoint-3000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e26d6088b3cba4720112b0132a302a5c942d74195dbc2a3eb65e03180472fcf
+size 433279996
diff --git a/run-3/checkpoint-3000/optimizer.pt b/run-3/checkpoint-3000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f314713207bdb6ca5f3f33fe1ee5fd5e0cab194d
--- /dev/null
+++ b/run-3/checkpoint-3000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80ed5d176426e957d86161b3e55b2b6ce0107a51eceddad1b88636534d9ae037
+size 866681082
diff --git a/run-3/checkpoint-3000/rng_state.pth b/run-3/checkpoint-3000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d717cb03d14eef32afe17820643b27cad107af13
--- /dev/null
+++ b/run-3/checkpoint-3000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a039281de77e6b9a31e662411886dca5f1d510240f4f6e504c9f52ef9b31a88b
+size 14244
diff --git a/run-3/checkpoint-3000/scheduler.pt b/run-3/checkpoint-3000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8d59179f2c37453561005d5c96abea3825b87060
--- /dev/null
+++ b/run-3/checkpoint-3000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adff77cc27f1038a8d6aff025d674cc722f885f789d53f585dab426dcbbeb171
+size 1064
diff --git a/run-3/checkpoint-3000/trainer_state.json b/run-3/checkpoint-3000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..17944facbb42fd63e447573ca9022afb2d05d42a
--- /dev/null
+++ b/run-3/checkpoint-3000/trainer_state.json
@@ -0,0 +1,852 @@
+{
+  "best_metric": 0.884020618556701,
+  "best_model_checkpoint": "bert_trainer/run-3/checkpoint-2750",
+  "epoch": 1.9342359767891684,
+  "eval_steps": 250,
+  "global_step": 3000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 1.3880483293934951e-05,
+      "loss": 1.3534,
+      "step": 25
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 1.0289,
+      "step": 50
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.8774,
+      "step": 75
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.7025,
+      "step": 100
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.8032,
+      "step": 125
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.808,
+      "step": 150
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.651,
+      "step": 175
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.6946,
+      "step": 200
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.6587,
+      "step": 225
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5824,
+      "step": 250
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.8112113402061856,
+      "eval_loss": 0.5165922045707703,
+      "eval_runtime": 30.3844,
+      "eval_samples_per_second": 102.158,
+      "eval_steps_per_second": 6.385,
+      "step": 250
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.546,
+      "step": 275
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5713,
+      "step": 300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4633,
+      "step": 325
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5839,
+      "step": 350
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4685,
+      "step": 375
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5632,
+      "step": 400
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.52,
+      "step": 425
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5565,
+      "step": 450
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5394,
+      "step": 475
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4944,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8244201030927835,
+      "eval_loss": 0.4521254003047943,
+      "eval_runtime": 30.375,
+      "eval_samples_per_second": 102.189,
+      "eval_steps_per_second": 6.387,
+      "step": 500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5233,
+      "step": 525
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.475,
+      "step": 550
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4169,
+      "step": 575
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4961,
+      "step": 600
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5372,
+      "step": 625
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5151,
+      "step": 650
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5491,
+      "step": 675
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4596,
+      "step": 700
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4722,
+      "step": 725
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3632,
+      "step": 750
+    },
+    {
+      "epoch": 0.48,
+      "eval_accuracy": 0.8344072164948454,
+      "eval_loss": 0.48342078924179077,
+      "eval_runtime": 30.4129,
+      "eval_samples_per_second": 102.062,
+      "eval_steps_per_second": 6.379,
+      "step": 750
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5665,
+      "step": 775
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3748,
+      "step": 800
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4397,
+      "step": 825
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4141,
+      "step": 850
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4029,
+      "step": 875
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4191,
+      "step": 900
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5094,
+      "step": 925
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4308,
+      "step": 950
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3921,
+      "step": 975
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4381,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8588917525773195,
+      "eval_loss": 0.4446374177932739,
+      "eval_runtime": 29.9606,
+      "eval_samples_per_second": 103.603,
+      "eval_steps_per_second": 6.475,
+      "step": 1000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4766,
+      "step": 1025
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4957,
+      "step": 1050
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4285,
+      "step": 1075
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3494,
+      "step": 1100
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4066,
+      "step": 1125
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3833,
+      "step": 1150
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4553,
+      "step": 1175
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4185,
+      "step": 1200
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3519,
+      "step": 1225
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4848,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "eval_accuracy": 0.8582474226804123,
+      "eval_loss": 0.408128023147583,
+      "eval_runtime": 29.8798,
+      "eval_samples_per_second": 103.883,
+      "eval_steps_per_second": 6.493,
+      "step": 1250
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.539,
+      "step": 1275
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3632,
+      "step": 1300
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3861,
+      "step": 1325
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.5181,
+      "step": 1350
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4158,
+      "step": 1375
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3584,
+      "step": 1400
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4162,
+      "step": 1425
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4607,
+      "step": 1450
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4146,
+      "step": 1475
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3954,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8714561855670103,
+      "eval_loss": 0.38059666752815247,
+      "eval_runtime": 29.8772,
+      "eval_samples_per_second": 103.892,
+      "eval_steps_per_second": 6.493,
+      "step": 1500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3881,
+      "step": 1525
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3825,
+      "step": 1550
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3082,
+      "step": 1575
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3367,
+      "step": 1600
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3191,
+      "step": 1625
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3298,
+      "step": 1650
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2631,
+      "step": 1675
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1897,
+      "step": 1700
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3901,
+      "step": 1725
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3268,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "eval_accuracy": 0.8714561855670103,
+      "eval_loss": 0.5426503419876099,
+      "eval_runtime": 29.922,
+      "eval_samples_per_second": 103.737,
+      "eval_steps_per_second": 6.484,
+      "step": 1750
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2771,
+      "step": 1775
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3346,
+      "step": 1800
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.312,
+      "step": 1825
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3482,
+      "step": 1850
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3186,
+      "step": 1875
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.317,
+      "step": 1900
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3214,
+      "step": 1925
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2073,
+      "step": 1950
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3709,
+      "step": 1975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2492,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8730670103092784,
+      "eval_loss": 0.4012953042984009,
+      "eval_runtime": 29.8764,
+      "eval_samples_per_second": 103.895,
+      "eval_steps_per_second": 6.493,
+      "step": 2000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2831,
+      "step": 2025
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2479,
+      "step": 2050
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2467,
+      "step": 2075
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.332,
+      "step": 2100
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2565,
+      "step": 2125
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4016,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1485,
+      "step": 2175
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3333,
+      "step": 2200
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3541,
+      "step": 2225
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1829,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "eval_accuracy": 0.8704896907216495,
+      "eval_loss": 0.49169790744781494,
+      "eval_runtime": 29.9217,
+      "eval_samples_per_second": 103.737,
+      "eval_steps_per_second": 6.484,
+      "step": 2250
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2888,
+      "step": 2275
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2472,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3547,
+      "step": 2325
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.1869,
+      "step": 2350
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3094,
+      "step": 2375
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.314,
+      "step": 2400
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3161,
+      "step": 2425
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3445,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.4138,
+      "step": 2475
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2214,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.852770618556701,
+      "eval_loss": 0.5358509421348572,
+      "eval_runtime": 29.9717,
+      "eval_samples_per_second": 103.564,
+      "eval_steps_per_second": 6.473,
+      "step": 2500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3897,
+      "step": 2525
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3371,
+      "step": 2550
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3278,
+      "step": 2575
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2226,
+      "step": 2600
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2836,
+      "step": 2625
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3236,
+      "step": 2650
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3125,
+      "step": 2675
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3267,
+      "step": 2700
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2427,
+      "step": 2725
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2973,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "eval_accuracy": 0.884020618556701,
+      "eval_loss": 0.42780932784080505,
+      "eval_runtime": 29.8873,
+      "eval_samples_per_second": 103.857,
+      "eval_steps_per_second": 6.491,
+      "step": 2750
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2346,
+      "step": 2775
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3316,
+      "step": 2800
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2187,
+      "step": 2825
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2961,
+      "step": 2850
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2899,
+      "step": 2875
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3272,
+      "step": 2900
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2203,
+      "step": 2925
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.2468,
+      "step": 2950
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3294,
+      "step": 2975
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2.7760966587869903e-05,
+      "loss": 0.3101,
+      "step": 3000
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8695231958762887,
+      "eval_loss": 0.4797918200492859,
+      "eval_runtime": 29.9807,
+      "eval_samples_per_second": 103.533,
+      "eval_steps_per_second": 6.471,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 250,
+  "total_flos": 6314309182851072.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 2.7760966587869903e-05,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 8
+  }
+}
diff --git a/run-3/checkpoint-3000/training_args.bin b/run-3/checkpoint-3000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..50cc4ec23d371b30ae6512db04b9c655f9dcecd3
--- /dev/null
+++ b/run-3/checkpoint-3000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2175af5e4c44e0db4ee2b9299c68887232a3ab86bd73b88fdec63aaf1669da81
+size 4728
diff --git a/run-4/checkpoint-2500/config.json b/run-4/checkpoint-2500/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-4/checkpoint-2500/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-4/checkpoint-2500/model.safetensors b/run-4/checkpoint-2500/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..764a3c76b17d0ffea12ca0c623fedbf064048988
--- /dev/null
+++ b/run-4/checkpoint-2500/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38e2b4a9a173b8acab929cb2ea204720af2cf0b255f90e55be522fb34c041a5e
+size 433279996
diff --git a/run-4/checkpoint-2500/optimizer.pt b/run-4/checkpoint-2500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7adcddbc1f5987df5fd46aeddcaee990244558ef
--- /dev/null
+++ b/run-4/checkpoint-2500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:545f26d60a5ff182494efe05a78b6cf5421600caa72c9185beb786836356493b
+size 866681082
diff --git a/run-4/checkpoint-2500/rng_state.pth b/run-4/checkpoint-2500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b68d4eaf43291f178fb309f1adb0a7b459dd867c
--- /dev/null
+++ b/run-4/checkpoint-2500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e254f06883fec1e30ea9be59e97a84e2814064af09a93a2e0e9f4655ea082cd
+size 14244
diff --git a/run-4/checkpoint-2500/scheduler.pt b/run-4/checkpoint-2500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9bc70f9430d850a7bb7555debaa663376c20362c
--- /dev/null
+++ b/run-4/checkpoint-2500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3aa13ef7864fffc27c66f8187bba19305ec4a345e5fb010f3e70efbb9f481b37
+size 1064
diff --git a/run-4/checkpoint-2500/trainer_state.json b/run-4/checkpoint-2500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..0f1d6e9979f2385d7d70e25b9c727d88e07f9fd1
--- /dev/null
+++ b/run-4/checkpoint-2500/trainer_state.json
@@ -0,0 +1,714 @@
+{
+  "best_metric": 0.8788659793814433,
+  "best_model_checkpoint": "bert_trainer/run-4/checkpoint-2000",
+  "epoch": 3.2216494845360826,
+  "eval_steps": 250,
+  "global_step": 2500,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 3.6540885491028537e-06,
+      "loss": 1.4368,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 1.1679,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 1.0475,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.9295,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.8447,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.7885,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.7324,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6723,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6116,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6183,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8028350515463918,
+      "eval_loss": 0.5710608959197998,
+      "eval_runtime": 70.1794,
+      "eval_samples_per_second": 44.229,
+      "eval_steps_per_second": 2.764,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5932,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5536,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6114,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6167,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.558,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5465,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.538,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5061,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5338,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4463,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8350515463917526,
+      "eval_loss": 0.46032479405403137,
+      "eval_runtime": 70.1578,
+      "eval_samples_per_second": 44.243,
+      "eval_steps_per_second": 2.765,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.537,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4766,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.425,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4592,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4481,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4451,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4429,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4401,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3966,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3764,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8547036082474226,
+      "eval_loss": 0.4088385999202728,
+      "eval_runtime": 70.1255,
+      "eval_samples_per_second": 44.263,
+      "eval_steps_per_second": 2.766,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4036,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.351,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3393,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3148,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3454,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3262,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3758,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3801,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3554,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3555,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8692010309278351,
+      "eval_loss": 0.3840981423854828,
+      "eval_runtime": 70.1528,
+      "eval_samples_per_second": 44.246,
+      "eval_steps_per_second": 2.765,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3394,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3371,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3528,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2884,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2866,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2951,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.325,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3478,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3645,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3301,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8521262886597938,
+      "eval_loss": 0.4127959907054901,
+      "eval_runtime": 70.9551,
+      "eval_samples_per_second": 43.746,
+      "eval_steps_per_second": 2.734,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4149,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3465,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2909,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3678,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2809,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.346,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3013,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2969,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2765,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2965,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8717783505154639,
+      "eval_loss": 0.3640108108520508,
+      "eval_runtime": 70.1759,
+      "eval_samples_per_second": 44.232,
+      "eval_steps_per_second": 2.764,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2653,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3222,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2863,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2854,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2044,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.255,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2318,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2424,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2679,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.213,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8772551546391752,
+      "eval_loss": 0.41377973556518555,
+      "eval_runtime": 70.1634,
+      "eval_samples_per_second": 44.24,
+      "eval_steps_per_second": 2.765,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2542,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2385,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1808,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.214,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1793,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2128,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2436,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2224,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1998,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1944,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8788659793814433,
+      "eval_loss": 0.3984367251396179,
+      "eval_runtime": 70.1229,
+      "eval_samples_per_second": 44.265,
+      "eval_steps_per_second": 2.767,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2178,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2343,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2365,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1938,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2154,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2159,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2322,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2187,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2133,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3084,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8663015463917526,
+      "eval_loss": 0.41974076628685,
+      "eval_runtime": 70.1419,
+      "eval_samples_per_second": 44.253,
+      "eval_steps_per_second": 2.766,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2541,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1746,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2357,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.0945,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1552,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1606,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1474,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1521,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1468,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1137,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.8788659793814433,
+      "eval_loss": 0.4812679588794708,
+      "eval_runtime": 70.1728,
+      "eval_samples_per_second": 44.234,
+      "eval_steps_per_second": 2.765,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.051683215428608e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 7.308177098205707e-06,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-4/checkpoint-2500/training_args.bin b/run-4/checkpoint-2500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6c7cb475dec463c51866e7796afcdd113cd2e8b9
--- /dev/null
+++ b/run-4/checkpoint-2500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:866e09946e4ff26c3eb6ee3eebf4634a20a78365e95d2455b97ebf938806bdf9
+size 4728
diff --git a/run-4/checkpoint-2750/config.json b/run-4/checkpoint-2750/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-4/checkpoint-2750/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-4/checkpoint-2750/model.safetensors b/run-4/checkpoint-2750/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e899289721183f787e33d8bc8a0e06d748d44932
--- /dev/null
+++ b/run-4/checkpoint-2750/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:486ca5c0c52964413206c1ba7362fa5fe2c9671505111fec7d6747031b5ebabd
+size 433279996
diff --git a/run-4/checkpoint-2750/optimizer.pt b/run-4/checkpoint-2750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..abc3600fa1b368b672b59c0059f326c7c38b8931
--- /dev/null
+++ b/run-4/checkpoint-2750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a91fa5a2133d9a8de8ac1ea0fc5aafc6ffbd52ee2406d901de1c93557543f90
+size 866681082
diff --git a/run-4/checkpoint-2750/rng_state.pth b/run-4/checkpoint-2750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..eccfea317f0c1eb50423f8ebdfd26baba83a270c
--- /dev/null
+++ b/run-4/checkpoint-2750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2f146112a2b6651d5740f314c7a7760a6b10b307030d3ca8f929af57ce20006
+size 14244
diff --git a/run-4/checkpoint-2750/scheduler.pt b/run-4/checkpoint-2750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d3dfafe392e19cbef22ac0c0b38a610057603ecf
--- /dev/null
+++ b/run-4/checkpoint-2750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22eba7af503e4ee52d91910c6a53cc8801228b83199f3004c41b5caee2b570b7
+size 1064
diff --git a/run-4/checkpoint-2750/trainer_state.json b/run-4/checkpoint-2750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..07ffbc7ebf99c1425a8c3c746b4605e548323dc2
--- /dev/null
+++ b/run-4/checkpoint-2750/trainer_state.json
@@ -0,0 +1,783 @@
+{
+  "best_metric": 0.8788659793814433,
+  "best_model_checkpoint": "bert_trainer/run-4/checkpoint-2000",
+  "epoch": 3.5438144329896906,
+  "eval_steps": 250,
+  "global_step": 2750,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 3.6540885491028537e-06,
+      "loss": 1.4368,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 1.1679,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 1.0475,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.9295,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.8447,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.7885,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.7324,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6723,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6116,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6183,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8028350515463918,
+      "eval_loss": 0.5710608959197998,
+      "eval_runtime": 70.1794,
+      "eval_samples_per_second": 44.229,
+      "eval_steps_per_second": 2.764,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5932,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5536,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6114,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6167,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.558,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5465,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.538,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5061,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5338,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4463,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8350515463917526,
+      "eval_loss": 0.46032479405403137,
+      "eval_runtime": 70.1578,
+      "eval_samples_per_second": 44.243,
+      "eval_steps_per_second": 2.765,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.537,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4766,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.425,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4592,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4481,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4451,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4429,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4401,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3966,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3764,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8547036082474226,
+      "eval_loss": 0.4088385999202728,
+      "eval_runtime": 70.1255,
+      "eval_samples_per_second": 44.263,
+      "eval_steps_per_second": 2.766,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4036,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.351,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3393,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3148,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3454,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3262,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3758,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3801,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3554,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3555,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8692010309278351,
+      "eval_loss": 0.3840981423854828,
+      "eval_runtime": 70.1528,
+      "eval_samples_per_second": 44.246,
+      "eval_steps_per_second": 2.765,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3394,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3371,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3528,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2884,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2866,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2951,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.325,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3478,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3645,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3301,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8521262886597938,
+      "eval_loss": 0.4127959907054901,
+      "eval_runtime": 70.9551,
+      "eval_samples_per_second": 43.746,
+      "eval_steps_per_second": 2.734,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4149,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3465,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2909,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3678,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2809,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.346,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3013,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2969,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2765,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2965,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8717783505154639,
+      "eval_loss": 0.3640108108520508,
+      "eval_runtime": 70.1759,
+      "eval_samples_per_second": 44.232,
+      "eval_steps_per_second": 2.764,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2653,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3222,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2863,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2854,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2044,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.255,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2318,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2424,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2679,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.213,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8772551546391752,
+      "eval_loss": 0.41377973556518555,
+      "eval_runtime": 70.1634,
+      "eval_samples_per_second": 44.24,
+      "eval_steps_per_second": 2.765,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2542,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2385,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1808,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.214,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1793,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2128,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2436,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2224,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1998,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1944,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8788659793814433,
+      "eval_loss": 0.3984367251396179,
+      "eval_runtime": 70.1229,
+      "eval_samples_per_second": 44.265,
+      "eval_steps_per_second": 2.767,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2178,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2343,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2365,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1938,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2154,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2159,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2322,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2187,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2133,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3084,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8663015463917526,
+      "eval_loss": 0.41974076628685,
+      "eval_runtime": 70.1419,
+      "eval_samples_per_second": 44.253,
+      "eval_steps_per_second": 2.766,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2541,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1746,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2357,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.0945,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1552,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1606,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1474,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1521,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1468,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1137,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.8788659793814433,
+      "eval_loss": 0.4812679588794708,
+      "eval_runtime": 70.1728,
+      "eval_samples_per_second": 44.234,
+      "eval_steps_per_second": 2.765,
+      "step": 2500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1419,
+      "step": 2525
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1269,
+      "step": 2550
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1658,
+      "step": 2575
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1867,
+      "step": 2600
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1526,
+      "step": 2625
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1581,
+      "step": 2650
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1578,
+      "step": 2675
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1197,
+      "step": 2700
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1238,
+      "step": 2725
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1939,
+      "step": 2750
+    },
+    {
+      "epoch": 3.54,
+      "eval_accuracy": 0.8624355670103093,
+      "eval_loss": 0.5447501540184021,
+      "eval_runtime": 70.1531,
+      "eval_samples_per_second": 44.246,
+      "eval_steps_per_second": 2.765,
+      "step": 2750
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.156930472414208e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 7.308177098205707e-06,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-4/checkpoint-2750/training_args.bin b/run-4/checkpoint-2750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6c7cb475dec463c51866e7796afcdd113cd2e8b9
--- /dev/null
+++ b/run-4/checkpoint-2750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:866e09946e4ff26c3eb6ee3eebf4634a20a78365e95d2455b97ebf938806bdf9
+size 4728
diff --git a/run-4/checkpoint-3000/config.json b/run-4/checkpoint-3000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-4/checkpoint-3000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-4/checkpoint-3000/model.safetensors b/run-4/checkpoint-3000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..30e6b20d1ae88d09befe62bb58cdb97a3b36e7e2
--- /dev/null
+++ b/run-4/checkpoint-3000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c21d389222553424951b7378043b4432e37b02ab80729327bfc65b032789334
+size 433279996
diff --git a/run-4/checkpoint-3000/optimizer.pt b/run-4/checkpoint-3000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..d4bc172c85daf720ef051a6b7486e12d3ca9720f
--- /dev/null
+++ b/run-4/checkpoint-3000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a104b92a9e5b64cb1e9ece8a7a20b08272a5cc3a3076ebda7e19b92b5c8aa6bf
+size 866681082
diff --git a/run-4/checkpoint-3000/rng_state.pth b/run-4/checkpoint-3000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..134d7b84b2bcb7c8af8198d9d691b9391bfa81d0
--- /dev/null
+++ b/run-4/checkpoint-3000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5475dcc749e5bb84d23e503506ffb30e2c214153e88098ffa2eba5466de55a64
+size 14244
diff --git a/run-4/checkpoint-3000/scheduler.pt b/run-4/checkpoint-3000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..969b98116995fe54b9fbf54a1be122c61a91281b
--- /dev/null
+++ b/run-4/checkpoint-3000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d1a53eb60f72a811e542c634ff3778420f241e816a8829f657b152fb1c35264
+size 1064
diff --git a/run-4/checkpoint-3000/trainer_state.json b/run-4/checkpoint-3000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..6723a6c35cb882a4869c8253d5ac759f6cae82d6
--- /dev/null
+++ b/run-4/checkpoint-3000/trainer_state.json
@@ -0,0 +1,852 @@
+{
+  "best_metric": 0.8830541237113402,
+  "best_model_checkpoint": "bert_trainer/run-4/checkpoint-3000",
+  "epoch": 3.865979381443299,
+  "eval_steps": 250,
+  "global_step": 3000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 3.6540885491028537e-06,
+      "loss": 1.4368,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 1.1679,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 1.0475,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.9295,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.8447,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.7885,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.7324,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6723,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6116,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6183,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.8028350515463918,
+      "eval_loss": 0.5710608959197998,
+      "eval_runtime": 70.1794,
+      "eval_samples_per_second": 44.229,
+      "eval_steps_per_second": 2.764,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5932,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5536,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6114,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.6167,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.558,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5465,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.538,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5061,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.5338,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4463,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8350515463917526,
+      "eval_loss": 0.46032479405403137,
+      "eval_runtime": 70.1578,
+      "eval_samples_per_second": 44.243,
+      "eval_steps_per_second": 2.765,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.537,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4766,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.425,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4592,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4481,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4451,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4429,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4401,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3966,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3764,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8547036082474226,
+      "eval_loss": 0.4088385999202728,
+      "eval_runtime": 70.1255,
+      "eval_samples_per_second": 44.263,
+      "eval_steps_per_second": 2.766,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4036,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.351,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3393,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3148,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3454,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3262,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3758,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3801,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3554,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3555,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8692010309278351,
+      "eval_loss": 0.3840981423854828,
+      "eval_runtime": 70.1528,
+      "eval_samples_per_second": 44.246,
+      "eval_steps_per_second": 2.765,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3394,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3371,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3528,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2884,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2866,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2951,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.325,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3478,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3645,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3301,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8521262886597938,
+      "eval_loss": 0.4127959907054901,
+      "eval_runtime": 70.9551,
+      "eval_samples_per_second": 43.746,
+      "eval_steps_per_second": 2.734,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.4149,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3465,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2909,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3678,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2809,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.346,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3013,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2969,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2765,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2965,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8717783505154639,
+      "eval_loss": 0.3640108108520508,
+      "eval_runtime": 70.1759,
+      "eval_samples_per_second": 44.232,
+      "eval_steps_per_second": 2.764,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2653,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3222,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2863,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2854,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2044,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.255,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2318,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2424,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2679,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.213,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8772551546391752,
+      "eval_loss": 0.41377973556518555,
+      "eval_runtime": 70.1634,
+      "eval_samples_per_second": 44.24,
+      "eval_steps_per_second": 2.765,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2542,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2385,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1808,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.214,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1793,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2128,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2436,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2224,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1998,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1944,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8788659793814433,
+      "eval_loss": 0.3984367251396179,
+      "eval_runtime": 70.1229,
+      "eval_samples_per_second": 44.265,
+      "eval_steps_per_second": 2.767,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2178,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2343,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2365,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1938,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2154,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2159,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2322,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2187,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2133,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.3084,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8663015463917526,
+      "eval_loss": 0.41974076628685,
+      "eval_runtime": 70.1419,
+      "eval_samples_per_second": 44.253,
+      "eval_steps_per_second": 2.766,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2541,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1746,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2357,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.0945,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1552,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1606,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1474,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1521,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1468,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1137,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.8788659793814433,
+      "eval_loss": 0.4812679588794708,
+      "eval_runtime": 70.1728,
+      "eval_samples_per_second": 44.234,
+      "eval_steps_per_second": 2.765,
+      "step": 2500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1419,
+      "step": 2525
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1269,
+      "step": 2550
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1658,
+      "step": 2575
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1867,
+      "step": 2600
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1526,
+      "step": 2625
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1581,
+      "step": 2650
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1578,
+      "step": 2675
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1197,
+      "step": 2700
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1238,
+      "step": 2725
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1939,
+      "step": 2750
+    },
+    {
+      "epoch": 3.54,
+      "eval_accuracy": 0.8624355670103093,
+      "eval_loss": 0.5447501540184021,
+      "eval_runtime": 70.1531,
+      "eval_samples_per_second": 44.246,
+      "eval_steps_per_second": 2.765,
+      "step": 2750
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1061,
+      "step": 2775
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1304,
+      "step": 2800
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1149,
+      "step": 2825
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.2885,
+      "step": 2850
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.0996,
+      "step": 2875
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1914,
+      "step": 2900
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1325,
+      "step": 2925
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.115,
+      "step": 2950
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1132,
+      "step": 2975
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 7.308177098205707e-06,
+      "loss": 0.1746,
+      "step": 3000
+    },
+    {
+      "epoch": 3.87,
+      "eval_accuracy": 0.8830541237113402,
+      "eval_loss": 0.5246909856796265,
+      "eval_runtime": 70.1829,
+      "eval_samples_per_second": 44.227,
+      "eval_steps_per_second": 2.764,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.262177729399808e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 7.308177098205707e-06,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-4/checkpoint-3000/training_args.bin b/run-4/checkpoint-3000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..6c7cb475dec463c51866e7796afcdd113cd2e8b9
--- /dev/null
+++ b/run-4/checkpoint-3000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:866e09946e4ff26c3eb6ee3eebf4634a20a78365e95d2455b97ebf938806bdf9
+size 4728
diff --git a/run-5/checkpoint-1000/config.json b/run-5/checkpoint-1000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-5/checkpoint-1000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-5/checkpoint-1000/model.safetensors b/run-5/checkpoint-1000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7ade02a7f503efe01f3885e22e55af5b01d0ce97
--- /dev/null
+++ b/run-5/checkpoint-1000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:469d9d7e0084db7398eba3dacd3f166f7f8e71c57b73cb964a58ee2688b73037
+size 433279996
diff --git a/run-5/checkpoint-1000/optimizer.pt b/run-5/checkpoint-1000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0a24a594dc0ccfc9351c7718405853e9f762d5e6
--- /dev/null
+++ b/run-5/checkpoint-1000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68d10ba89e115ecac8b0d07a5467069a7c72e5e4cac8c8ec5f18dab8c19c2047
+size 866681082
diff --git a/run-5/checkpoint-1000/rng_state.pth b/run-5/checkpoint-1000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..45858f484cdbd5c8a3b0751391a982be12a141a2
--- /dev/null
+++ b/run-5/checkpoint-1000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f127fb2222d61995ad048f04448774ec877fc92b6c22fe18f0a1e0d3e5e9d66
+size 14244
diff --git a/run-5/checkpoint-1000/scheduler.pt b/run-5/checkpoint-1000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f6d53b17c9894e340ee6980d64f5c494464af7e7
--- /dev/null
+++ b/run-5/checkpoint-1000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b42b5358e79bac2e4f3421f572886fd62aeabc2b0997925b0793c7cd8f8ab88
+size 1064
diff --git a/run-5/checkpoint-1000/trainer_state.json b/run-5/checkpoint-1000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..384d4276692a107a83b078a8861fd59896f7b0e6
--- /dev/null
+++ b/run-5/checkpoint-1000/trainer_state.json
@@ -0,0 +1,300 @@
+{
+  "best_metric": 0.8701675257731959,
+  "best_model_checkpoint": "bert_trainer/run-5/checkpoint-1000",
+  "epoch": 1.2886597938144329,
+  "eval_steps": 250,
+  "global_step": 1000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.467798096290063e-05,
+      "loss": 1.272,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.8731,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.7963,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.7036,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.6194,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.6329,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5739,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5285,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5168,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4854,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.821520618556701,
+      "eval_loss": 0.4599289894104004,
+      "eval_runtime": 58.688,
+      "eval_samples_per_second": 52.89,
+      "eval_steps_per_second": 13.222,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4925,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4808,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5584,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5209,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4384,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4718,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4559,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4387,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4696,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3684,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8563144329896907,
+      "eval_loss": 0.4029282331466675,
+      "eval_runtime": 58.7278,
+      "eval_samples_per_second": 52.854,
+      "eval_steps_per_second": 13.214,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4881,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4714,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3966,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3674,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4573,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4286,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4245,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.449,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4077,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3337,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8653350515463918,
+      "eval_loss": 0.38506096601486206,
+      "eval_runtime": 61.6758,
+      "eval_samples_per_second": 50.328,
+      "eval_steps_per_second": 12.582,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3939,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3132,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2318,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2247,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3177,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2383,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.264,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2581,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2401,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3105,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8701675257731959,
+      "eval_loss": 0.4053144156932831,
+      "eval_runtime": 29.7918,
+      "eval_samples_per_second": 104.19,
+      "eval_steps_per_second": 26.047,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 4207259097999360.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 4.935596192580126e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-5/checkpoint-1000/training_args.bin b/run-5/checkpoint-1000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8c0bea560e0ecbe2e0ce61eade6c983ffb635abd
--- /dev/null
+++ b/run-5/checkpoint-1000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd3e53a63a07d35b153ee5aa4f711ac963c5d7f1200ab355158bc4f46fde13a2
+size 4728
diff --git a/run-5/checkpoint-1250/config.json b/run-5/checkpoint-1250/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-5/checkpoint-1250/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-5/checkpoint-1250/model.safetensors b/run-5/checkpoint-1250/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6a5de2c2c14801cae8654da2aa2358a72df67246
--- /dev/null
+++ b/run-5/checkpoint-1250/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17e888c37dcce0a7c69cc995922daebe7a6e90ffc9c6f465e4c461d7de9437fa
+size 433279996
diff --git a/run-5/checkpoint-1250/optimizer.pt b/run-5/checkpoint-1250/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9c98dd85f12ba29682c62bd91cf9cea9cf8fe096
--- /dev/null
+++ b/run-5/checkpoint-1250/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59c7fb910d6356b0734921d203f1f3aca5a2f2c1fa0dd1570ca034b745955204
+size 866681082
diff --git a/run-5/checkpoint-1250/rng_state.pth b/run-5/checkpoint-1250/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3bb44747e763c2dcc401f44214da0576e4e803e4
--- /dev/null
+++ b/run-5/checkpoint-1250/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7574999ed41e46ff3f6ab95ff2dff5798096013a38c58ca815181da639e6399c
+size 14244
diff --git a/run-5/checkpoint-1250/scheduler.pt b/run-5/checkpoint-1250/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..cb419694d665e2f671d104f48c4261b32ba50070
--- /dev/null
+++ b/run-5/checkpoint-1250/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d54350316be87d12d65b9d4c64ff7b6c9cc504b19f56b223d0afb962580b80c
+size 1064
diff --git a/run-5/checkpoint-1250/trainer_state.json b/run-5/checkpoint-1250/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7ad87b650d4078a1e3597ca68056bcb35a7b8e9f
--- /dev/null
+++ b/run-5/checkpoint-1250/trainer_state.json
@@ -0,0 +1,369 @@
+{
+  "best_metric": 0.8701675257731959,
+  "best_model_checkpoint": "bert_trainer/run-5/checkpoint-1000",
+  "epoch": 1.6108247422680413,
+  "eval_steps": 250,
+  "global_step": 1250,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.467798096290063e-05,
+      "loss": 1.272,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.8731,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.7963,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.7036,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.6194,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.6329,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5739,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5285,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5168,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4854,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.821520618556701,
+      "eval_loss": 0.4599289894104004,
+      "eval_runtime": 58.688,
+      "eval_samples_per_second": 52.89,
+      "eval_steps_per_second": 13.222,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4925,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4808,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5584,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5209,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4384,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4718,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4559,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4387,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4696,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3684,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8563144329896907,
+      "eval_loss": 0.4029282331466675,
+      "eval_runtime": 58.7278,
+      "eval_samples_per_second": 52.854,
+      "eval_steps_per_second": 13.214,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4881,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4714,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3966,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3674,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4573,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4286,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4245,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.449,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4077,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3337,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8653350515463918,
+      "eval_loss": 0.38506096601486206,
+      "eval_runtime": 61.6758,
+      "eval_samples_per_second": 50.328,
+      "eval_steps_per_second": 12.582,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3939,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3132,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2318,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2247,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3177,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2383,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.264,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2581,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2401,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3105,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8701675257731959,
+      "eval_loss": 0.4053144156932831,
+      "eval_runtime": 29.7918,
+      "eval_samples_per_second": 104.19,
+      "eval_steps_per_second": 26.047,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.267,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2586,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2878,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2126,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2285,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2599,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2691,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2536,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2712,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.2093,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8427835051546392,
+      "eval_loss": 0.4850337505340576,
+      "eval_runtime": 29.7884,
+      "eval_samples_per_second": 104.202,
+      "eval_steps_per_second": 26.05,
+      "step": 1250
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 5259731667855360.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 4.935596192580126e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-5/checkpoint-1250/training_args.bin b/run-5/checkpoint-1250/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8c0bea560e0ecbe2e0ce61eade6c983ffb635abd
--- /dev/null
+++ b/run-5/checkpoint-1250/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd3e53a63a07d35b153ee5aa4f711ac963c5d7f1200ab355158bc4f46fde13a2
+size 4728
diff --git a/run-5/checkpoint-750/config.json b/run-5/checkpoint-750/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-5/checkpoint-750/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-5/checkpoint-750/model.safetensors b/run-5/checkpoint-750/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..00c8c55e6afff41ae1698a0bb3817351a8e97814
--- /dev/null
+++ b/run-5/checkpoint-750/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30a670a2662b0f3604ed2cdb9ae2df14abe44a59d567d23dd388ba6319780b42
+size 433279996
diff --git a/run-5/checkpoint-750/optimizer.pt b/run-5/checkpoint-750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..cf740e85a4a4089515954532f438c36cbfedcdf4
--- /dev/null
+++ b/run-5/checkpoint-750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48660ab89cc1dea1ab3d1a5771920a6fa536265fbd831ea59cdd80d97f56fbe2
+size 866681082
diff --git a/run-5/checkpoint-750/rng_state.pth b/run-5/checkpoint-750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6407065aa4db83f99f5a45763c8e9df6113317dc
--- /dev/null
+++ b/run-5/checkpoint-750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:313bcb5b7ecc55139fd93d2a4ca526db70984f77ae2f25de8ef8161d3c44ab0a
+size 14244
diff --git a/run-5/checkpoint-750/scheduler.pt b/run-5/checkpoint-750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..778c7a5738b4abf10125b5aeb8f1d9c4772492f4
--- /dev/null
+++ b/run-5/checkpoint-750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58f6c584236f9863ac0de5373e8e0869d7ad3a7ba953a14d8392d38517f41389
+size 1064
diff --git a/run-5/checkpoint-750/trainer_state.json b/run-5/checkpoint-750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..aba08aba2e2ebd269a77355d84c47aeb29e79f63
--- /dev/null
+++ b/run-5/checkpoint-750/trainer_state.json
@@ -0,0 +1,231 @@
+{
+  "best_metric": 0.8653350515463918,
+  "best_model_checkpoint": "bert_trainer/run-5/checkpoint-750",
+  "epoch": 0.9664948453608248,
+  "eval_steps": 250,
+  "global_step": 750,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 2.467798096290063e-05,
+      "loss": 1.272,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.8731,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.7963,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.7036,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.6194,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.6329,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5739,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5285,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5168,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4854,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.821520618556701,
+      "eval_loss": 0.4599289894104004,
+      "eval_runtime": 58.688,
+      "eval_samples_per_second": 52.89,
+      "eval_steps_per_second": 13.222,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4925,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4808,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5584,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.5209,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4384,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4718,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4559,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4387,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4696,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3684,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.8563144329896907,
+      "eval_loss": 0.4029282331466675,
+      "eval_runtime": 58.7278,
+      "eval_samples_per_second": 52.854,
+      "eval_steps_per_second": 13.214,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4881,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4714,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3966,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3674,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4573,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4286,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4245,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.449,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.4077,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.935596192580126e-05,
+      "loss": 0.3337,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.8653350515463918,
+      "eval_loss": 0.38506096601486206,
+      "eval_runtime": 61.6758,
+      "eval_samples_per_second": 50.328,
+      "eval_steps_per_second": 12.582,
+      "step": 750
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 3157417709568000.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 4.935596192580126e-05,
+    "per_device_eval_batch_size": 4,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-5/checkpoint-750/training_args.bin b/run-5/checkpoint-750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8c0bea560e0ecbe2e0ce61eade6c983ffb635abd
--- /dev/null
+++ b/run-5/checkpoint-750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd3e53a63a07d35b153ee5aa4f711ac963c5d7f1200ab355158bc4f46fde13a2
+size 4728
diff --git a/run-6/checkpoint-2500/config.json b/run-6/checkpoint-2500/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-6/checkpoint-2500/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-6/checkpoint-2500/model.safetensors b/run-6/checkpoint-2500/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..accb6d03c4d3cef1ba5a4dd5395877b3c2ef7dd6
--- /dev/null
+++ b/run-6/checkpoint-2500/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:386c0ee3a33be4c3207118b31663fb490c5dde725ae1eecc9129ccdc761dc364
+size 433279996
diff --git a/run-6/checkpoint-2500/optimizer.pt b/run-6/checkpoint-2500/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..8e064d4b966dbba2185c703a158030c1b58d8ccc
--- /dev/null
+++ b/run-6/checkpoint-2500/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8136063276e926d813f88666638eb94f312fca88c2aac4e3860e2ea4e96c6b82
+size 866681082
diff --git a/run-6/checkpoint-2500/rng_state.pth b/run-6/checkpoint-2500/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..b68d4eaf43291f178fb309f1adb0a7b459dd867c
--- /dev/null
+++ b/run-6/checkpoint-2500/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e254f06883fec1e30ea9be59e97a84e2814064af09a93a2e0e9f4655ea082cd
+size 14244
diff --git a/run-6/checkpoint-2500/scheduler.pt b/run-6/checkpoint-2500/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..3e82d9742c1f67b44f780251a8414729b2e74028
--- /dev/null
+++ b/run-6/checkpoint-2500/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb748be919948975ed25907f83de9335eb0ccfb462db0e334f6124c2b473bfdd
+size 1064
diff --git a/run-6/checkpoint-2500/trainer_state.json b/run-6/checkpoint-2500/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..12906e4365a24c517bec62fd5a77daa3a76837c2
--- /dev/null
+++ b/run-6/checkpoint-2500/trainer_state.json
@@ -0,0 +1,714 @@
+{
+  "best_metric": 0.8353737113402062,
+  "best_model_checkpoint": "bert_trainer/run-6/checkpoint-2500",
+  "epoch": 3.2216494845360826,
+  "eval_steps": 250,
+  "global_step": 2500,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.985683583902487e-07,
+      "loss": 1.498,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.4082,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.2917,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.2083,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.1513,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.1182,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0869,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0574,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9843,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0196,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.6401417525773195,
+      "eval_loss": 0.9714874029159546,
+      "eval_runtime": 53.062,
+      "eval_samples_per_second": 58.498,
+      "eval_steps_per_second": 3.656,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9886,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9861,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9746,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9386,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9046,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9213,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.892,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8545,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.87,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8463,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.6978092783505154,
+      "eval_loss": 0.810293436050415,
+      "eval_runtime": 53.0534,
+      "eval_samples_per_second": 58.507,
+      "eval_steps_per_second": 3.657,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8516,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8086,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7894,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7754,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7558,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7155,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7151,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.758,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6958,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.68,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.7503221649484536,
+      "eval_loss": 0.6983952522277832,
+      "eval_runtime": 57.8438,
+      "eval_samples_per_second": 53.662,
+      "eval_steps_per_second": 3.354,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6553,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6829,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7116,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6283,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6559,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6118,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6675,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6426,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6344,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6754,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.7970360824742269,
+      "eval_loss": 0.6213943362236023,
+      "eval_runtime": 57.838,
+      "eval_samples_per_second": 53.667,
+      "eval_steps_per_second": 3.354,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6535,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.64,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6442,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5873,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.587,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5789,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.605,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6078,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6431,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5824,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.803479381443299,
+      "eval_loss": 0.5681772232055664,
+      "eval_runtime": 57.809,
+      "eval_samples_per_second": 53.694,
+      "eval_steps_per_second": 3.356,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.648,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5709,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5504,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5699,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5575,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5644,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.587,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5094,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5724,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5652,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8128221649484536,
+      "eval_loss": 0.5369997024536133,
+      "eval_runtime": 57.8152,
+      "eval_samples_per_second": 53.688,
+      "eval_steps_per_second": 3.356,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5197,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5499,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.58,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5348,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4741,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5582,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5219,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4802,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5147,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4912,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8234536082474226,
+      "eval_loss": 0.5059388279914856,
+      "eval_runtime": 57.8239,
+      "eval_samples_per_second": 53.68,
+      "eval_steps_per_second": 3.355,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5133,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4629,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4696,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4623,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4649,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5314,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4974,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.464,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5332,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4694,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8286082474226805,
+      "eval_loss": 0.48415493965148926,
+      "eval_runtime": 57.8531,
+      "eval_samples_per_second": 53.653,
+      "eval_steps_per_second": 3.353,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5191,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4969,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4499,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5144,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.482,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4193,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4928,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4515,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4775,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5456,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.833118556701031,
+      "eval_loss": 0.4647773802280426,
+      "eval_runtime": 57.8252,
+      "eval_samples_per_second": 53.679,
+      "eval_steps_per_second": 3.355,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5226,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4159,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4486,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3866,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4634,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4179,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4446,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4606,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4261,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3907,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.8353737113402062,
+      "eval_loss": 0.45589184761047363,
+      "eval_runtime": 57.8334,
+      "eval_samples_per_second": 53.671,
+      "eval_steps_per_second": 3.354,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.051683215428608e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.1971367167804974e-06,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-6/checkpoint-2500/training_args.bin b/run-6/checkpoint-2500/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..fb68c4d48e2b11b9c1da4d8eda6ee91004244ec7
--- /dev/null
+++ b/run-6/checkpoint-2500/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04a33255eeb366639c63b8cb3420b4a0657a70d449a809b3f07061def559312b
+size 4728
diff --git a/run-6/checkpoint-2750/config.json b/run-6/checkpoint-2750/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-6/checkpoint-2750/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-6/checkpoint-2750/model.safetensors b/run-6/checkpoint-2750/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..45157545903a0060e30153929d2781dda15a3adb
--- /dev/null
+++ b/run-6/checkpoint-2750/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11baed352187d9b2cc06893e83e1e10a8177e636488900626143359548936a66
+size 433279996
diff --git a/run-6/checkpoint-2750/optimizer.pt b/run-6/checkpoint-2750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c5fa823b28e4915d9f797d71e5a437c68bfc51f3
--- /dev/null
+++ b/run-6/checkpoint-2750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3588b421761e3f3cbe03fc46841f87bc2094d3cbf4c740c11195fe29a3cb99eb
+size 866681082
diff --git a/run-6/checkpoint-2750/rng_state.pth b/run-6/checkpoint-2750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..eccfea317f0c1eb50423f8ebdfd26baba83a270c
--- /dev/null
+++ b/run-6/checkpoint-2750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2f146112a2b6651d5740f314c7a7760a6b10b307030d3ca8f929af57ce20006
+size 14244
diff --git a/run-6/checkpoint-2750/scheduler.pt b/run-6/checkpoint-2750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4b02ecc615c41e3fdb185bbdf403961fac2c5bca
--- /dev/null
+++ b/run-6/checkpoint-2750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4e3e33a7532894bf764d3712f50f39409ac119fe6cbb287875a78200f1486c1
+size 1064
diff --git a/run-6/checkpoint-2750/trainer_state.json b/run-6/checkpoint-2750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..8278e5a63559fc88e409176e0b1177a9b9d283a9
--- /dev/null
+++ b/run-6/checkpoint-2750/trainer_state.json
@@ -0,0 +1,783 @@
+{
+  "best_metric": 0.8472938144329897,
+  "best_model_checkpoint": "bert_trainer/run-6/checkpoint-2750",
+  "epoch": 3.5438144329896906,
+  "eval_steps": 250,
+  "global_step": 2750,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.985683583902487e-07,
+      "loss": 1.498,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.4082,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.2917,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.2083,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.1513,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.1182,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0869,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0574,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9843,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0196,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.6401417525773195,
+      "eval_loss": 0.9714874029159546,
+      "eval_runtime": 53.062,
+      "eval_samples_per_second": 58.498,
+      "eval_steps_per_second": 3.656,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9886,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9861,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9746,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9386,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9046,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9213,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.892,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8545,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.87,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8463,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.6978092783505154,
+      "eval_loss": 0.810293436050415,
+      "eval_runtime": 53.0534,
+      "eval_samples_per_second": 58.507,
+      "eval_steps_per_second": 3.657,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8516,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8086,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7894,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7754,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7558,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7155,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7151,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.758,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6958,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.68,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.7503221649484536,
+      "eval_loss": 0.6983952522277832,
+      "eval_runtime": 57.8438,
+      "eval_samples_per_second": 53.662,
+      "eval_steps_per_second": 3.354,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6553,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6829,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7116,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6283,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6559,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6118,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6675,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6426,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6344,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6754,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.7970360824742269,
+      "eval_loss": 0.6213943362236023,
+      "eval_runtime": 57.838,
+      "eval_samples_per_second": 53.667,
+      "eval_steps_per_second": 3.354,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6535,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.64,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6442,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5873,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.587,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5789,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.605,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6078,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6431,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5824,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.803479381443299,
+      "eval_loss": 0.5681772232055664,
+      "eval_runtime": 57.809,
+      "eval_samples_per_second": 53.694,
+      "eval_steps_per_second": 3.356,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.648,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5709,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5504,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5699,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5575,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5644,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.587,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5094,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5724,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5652,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8128221649484536,
+      "eval_loss": 0.5369997024536133,
+      "eval_runtime": 57.8152,
+      "eval_samples_per_second": 53.688,
+      "eval_steps_per_second": 3.356,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5197,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5499,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.58,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5348,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4741,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5582,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5219,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4802,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5147,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4912,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8234536082474226,
+      "eval_loss": 0.5059388279914856,
+      "eval_runtime": 57.8239,
+      "eval_samples_per_second": 53.68,
+      "eval_steps_per_second": 3.355,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5133,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4629,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4696,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4623,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4649,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5314,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4974,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.464,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5332,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4694,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8286082474226805,
+      "eval_loss": 0.48415493965148926,
+      "eval_runtime": 57.8531,
+      "eval_samples_per_second": 53.653,
+      "eval_steps_per_second": 3.353,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5191,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4969,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4499,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5144,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.482,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4193,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4928,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4515,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4775,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5456,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.833118556701031,
+      "eval_loss": 0.4647773802280426,
+      "eval_runtime": 57.8252,
+      "eval_samples_per_second": 53.679,
+      "eval_steps_per_second": 3.355,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5226,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4159,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4486,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3866,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4634,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4179,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4446,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4606,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4261,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3907,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.8353737113402062,
+      "eval_loss": 0.45589184761047363,
+      "eval_runtime": 57.8334,
+      "eval_samples_per_second": 53.671,
+      "eval_steps_per_second": 3.354,
+      "step": 2500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3983,
+      "step": 2525
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4491,
+      "step": 2550
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4599,
+      "step": 2575
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4581,
+      "step": 2600
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4318,
+      "step": 2625
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4607,
+      "step": 2650
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5343,
+      "step": 2675
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.437,
+      "step": 2700
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3962,
+      "step": 2725
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4803,
+      "step": 2750
+    },
+    {
+      "epoch": 3.54,
+      "eval_accuracy": 0.8472938144329897,
+      "eval_loss": 0.43914180994033813,
+      "eval_runtime": 57.8467,
+      "eval_samples_per_second": 53.659,
+      "eval_steps_per_second": 3.354,
+      "step": 2750
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.156930472414208e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.1971367167804974e-06,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-6/checkpoint-2750/training_args.bin b/run-6/checkpoint-2750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..fb68c4d48e2b11b9c1da4d8eda6ee91004244ec7
--- /dev/null
+++ b/run-6/checkpoint-2750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04a33255eeb366639c63b8cb3420b4a0657a70d449a809b3f07061def559312b
+size 4728
diff --git a/run-6/checkpoint-3000/config.json b/run-6/checkpoint-3000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-6/checkpoint-3000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-6/checkpoint-3000/model.safetensors b/run-6/checkpoint-3000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ffd8c01d5fab941cdbd616013853960b697e2a12
--- /dev/null
+++ b/run-6/checkpoint-3000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9574990c14799882e496a412cc1ad7930b3d27a56a724d332786042f8f279990
+size 433279996
diff --git a/run-6/checkpoint-3000/optimizer.pt b/run-6/checkpoint-3000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..63946cc0fa7ab411f4eada51b3c88417ea0ad5d3
--- /dev/null
+++ b/run-6/checkpoint-3000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85a60fdadebd33b9befa0b30d21b1b3d20c4dd5ba9d5ea6787614d9eabc4ddeb
+size 866681082
diff --git a/run-6/checkpoint-3000/rng_state.pth b/run-6/checkpoint-3000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..134d7b84b2bcb7c8af8198d9d691b9391bfa81d0
--- /dev/null
+++ b/run-6/checkpoint-3000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5475dcc749e5bb84d23e503506ffb30e2c214153e88098ffa2eba5466de55a64
+size 14244
diff --git a/run-6/checkpoint-3000/scheduler.pt b/run-6/checkpoint-3000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..863a9e14258474d76125b72077b62e80569fd25c
--- /dev/null
+++ b/run-6/checkpoint-3000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba07881e6660efa5f6b75d7be0a56065646b8e0d831a6642180986b9ae5cdd1f
+size 1064
diff --git a/run-6/checkpoint-3000/trainer_state.json b/run-6/checkpoint-3000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..0ea62167e49da0585fe48cbd9ba646c26b931194
--- /dev/null
+++ b/run-6/checkpoint-3000/trainer_state.json
@@ -0,0 +1,852 @@
+{
+  "best_metric": 0.8472938144329897,
+  "best_model_checkpoint": "bert_trainer/run-6/checkpoint-2750",
+  "epoch": 3.865979381443299,
+  "eval_steps": 250,
+  "global_step": 3000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 5.985683583902487e-07,
+      "loss": 1.498,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.4082,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.2917,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.2083,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.1513,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.1182,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0869,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0574,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9843,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 1.0196,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.6401417525773195,
+      "eval_loss": 0.9714874029159546,
+      "eval_runtime": 53.062,
+      "eval_samples_per_second": 58.498,
+      "eval_steps_per_second": 3.656,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9886,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9861,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9746,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9386,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9046,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.9213,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.892,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8545,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.87,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8463,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.6978092783505154,
+      "eval_loss": 0.810293436050415,
+      "eval_runtime": 53.0534,
+      "eval_samples_per_second": 58.507,
+      "eval_steps_per_second": 3.657,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8516,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.8086,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7894,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7754,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7558,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7155,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7151,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.758,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6958,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.68,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.7503221649484536,
+      "eval_loss": 0.6983952522277832,
+      "eval_runtime": 57.8438,
+      "eval_samples_per_second": 53.662,
+      "eval_steps_per_second": 3.354,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6553,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6829,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.7116,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6283,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6559,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6118,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6675,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6426,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6344,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6754,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.7970360824742269,
+      "eval_loss": 0.6213943362236023,
+      "eval_runtime": 57.838,
+      "eval_samples_per_second": 53.667,
+      "eval_steps_per_second": 3.354,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6535,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.64,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6442,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5873,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.587,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5789,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.605,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6078,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.6431,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5824,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.803479381443299,
+      "eval_loss": 0.5681772232055664,
+      "eval_runtime": 57.809,
+      "eval_samples_per_second": 53.694,
+      "eval_steps_per_second": 3.356,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.648,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5709,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5504,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5699,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5575,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5644,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.587,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5094,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5724,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5652,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8128221649484536,
+      "eval_loss": 0.5369997024536133,
+      "eval_runtime": 57.8152,
+      "eval_samples_per_second": 53.688,
+      "eval_steps_per_second": 3.356,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5197,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5499,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.58,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5348,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4741,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5582,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5219,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4802,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5147,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4912,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8234536082474226,
+      "eval_loss": 0.5059388279914856,
+      "eval_runtime": 57.8239,
+      "eval_samples_per_second": 53.68,
+      "eval_steps_per_second": 3.355,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5133,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4629,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4696,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4623,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4649,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5314,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4974,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.464,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5332,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4694,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8286082474226805,
+      "eval_loss": 0.48415493965148926,
+      "eval_runtime": 57.8531,
+      "eval_samples_per_second": 53.653,
+      "eval_steps_per_second": 3.353,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5191,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4969,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4499,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5144,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.482,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4193,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4928,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4515,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4775,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5456,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.833118556701031,
+      "eval_loss": 0.4647773802280426,
+      "eval_runtime": 57.8252,
+      "eval_samples_per_second": 53.679,
+      "eval_steps_per_second": 3.355,
+      "step": 2250
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5226,
+      "step": 2275
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4159,
+      "step": 2300
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4486,
+      "step": 2325
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3866,
+      "step": 2350
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4634,
+      "step": 2375
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4179,
+      "step": 2400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4446,
+      "step": 2425
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4606,
+      "step": 2450
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4261,
+      "step": 2475
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3907,
+      "step": 2500
+    },
+    {
+      "epoch": 3.22,
+      "eval_accuracy": 0.8353737113402062,
+      "eval_loss": 0.45589184761047363,
+      "eval_runtime": 57.8334,
+      "eval_samples_per_second": 53.671,
+      "eval_steps_per_second": 3.354,
+      "step": 2500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3983,
+      "step": 2525
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4491,
+      "step": 2550
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4599,
+      "step": 2575
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4581,
+      "step": 2600
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4318,
+      "step": 2625
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4607,
+      "step": 2650
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5343,
+      "step": 2675
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.437,
+      "step": 2700
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3962,
+      "step": 2725
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4803,
+      "step": 2750
+    },
+    {
+      "epoch": 3.54,
+      "eval_accuracy": 0.8472938144329897,
+      "eval_loss": 0.43914180994033813,
+      "eval_runtime": 57.8467,
+      "eval_samples_per_second": 53.659,
+      "eval_steps_per_second": 3.354,
+      "step": 2750
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3769,
+      "step": 2775
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3426,
+      "step": 2800
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.419,
+      "step": 2825
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.5138,
+      "step": 2850
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3962,
+      "step": 2875
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3931,
+      "step": 2900
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3926,
+      "step": 2925
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3425,
+      "step": 2950
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.3612,
+      "step": 2975
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 1.1971367167804974e-06,
+      "loss": 0.4456,
+      "step": 3000
+    },
+    {
+      "epoch": 3.87,
+      "eval_accuracy": 0.8392396907216495,
+      "eval_loss": 0.44415542483329773,
+      "eval_runtime": 57.8667,
+      "eval_samples_per_second": 53.641,
+      "eval_steps_per_second": 3.353,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 1.262177729399808e+16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.1971367167804974e-06,
+    "per_device_eval_batch_size": 16,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-6/checkpoint-3000/training_args.bin b/run-6/checkpoint-3000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..fb68c4d48e2b11b9c1da4d8eda6ee91004244ec7
--- /dev/null
+++ b/run-6/checkpoint-3000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04a33255eeb366639c63b8cb3420b4a0657a70d449a809b3f07061def559312b
+size 4728
diff --git a/run-9/checkpoint-1750/config.json b/run-9/checkpoint-1750/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-9/checkpoint-1750/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-9/checkpoint-1750/model.safetensors b/run-9/checkpoint-1750/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f14b15ba1968e6f7625163db299367b497927c8b
--- /dev/null
+++ b/run-9/checkpoint-1750/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5283073d957e54f92c3388d7c565ed89b4aacc6f9cb3110fd1c50384f297e39c
+size 433279996
diff --git a/run-9/checkpoint-1750/optimizer.pt b/run-9/checkpoint-1750/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..402202894b27724775aecaff0e5a1e721259b4ff
--- /dev/null
+++ b/run-9/checkpoint-1750/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e71068dadbfb72aa6c89509c5fb418d3a6e0f50c08d32aea0194db46019803e
+size 866681082
diff --git a/run-9/checkpoint-1750/rng_state.pth b/run-9/checkpoint-1750/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..365c9213082b250ded7fb86003f89dc68f396aad
--- /dev/null
+++ b/run-9/checkpoint-1750/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57a5c388b76bb6f1ac90bbf9bb082c12e6642eaa83617e4fc9e06ed5e1cf1545
+size 14244
diff --git a/run-9/checkpoint-1750/scheduler.pt b/run-9/checkpoint-1750/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..1f0d55f356a4385451b5988b183c827ecac36c9d
--- /dev/null
+++ b/run-9/checkpoint-1750/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f63639b497a41b1bcb09c1c9b441f0b2f736e46c40294767dd582f2f08db4e4d
+size 1064
diff --git a/run-9/checkpoint-1750/trainer_state.json b/run-9/checkpoint-1750/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..b2adb5c983d1366fbf6e435e5e69d4386499e54f
--- /dev/null
+++ b/run-9/checkpoint-1750/trainer_state.json
@@ -0,0 +1,507 @@
+{
+  "best_metric": 0.8392396907216495,
+  "best_model_checkpoint": "bert_trainer/run-9/checkpoint-1750",
+  "epoch": 2.2551546391752577,
+  "eval_steps": 250,
+  "global_step": 1750,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.145788393269186e-07,
+      "loss": 1.4915,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.3598,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.236,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.1509,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0855,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0437,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0172,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9849,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9064,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9487,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.6591494845360825,
+      "eval_loss": 0.8960067629814148,
+      "eval_runtime": 122.5176,
+      "eval_samples_per_second": 25.335,
+      "eval_steps_per_second": 3.167,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8996,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9016,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8801,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8536,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7811,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7948,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7839,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7311,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7656,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7055,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.7625644329896907,
+      "eval_loss": 0.6912322640419006,
+      "eval_runtime": 122.6001,
+      "eval_samples_per_second": 25.318,
+      "eval_steps_per_second": 3.165,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7169,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6942,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6771,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6848,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6452,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5975,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6137,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6439,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.583,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5757,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.7976804123711341,
+      "eval_loss": 0.6034093499183655,
+      "eval_runtime": 122.5716,
+      "eval_samples_per_second": 25.324,
+      "eval_steps_per_second": 3.165,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5678,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5788,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6069,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5188,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.541,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5249,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5842,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5603,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5493,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5698,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8115335051546392,
+      "eval_loss": 0.5344901084899902,
+      "eval_runtime": 122.5922,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5672,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5462,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5508,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4896,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4862,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4929,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5232,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5364,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5632,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5132,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8273195876288659,
+      "eval_loss": 0.48882216215133667,
+      "eval_runtime": 122.5903,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.576,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4969,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4696,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4845,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4698,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4927,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5179,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4551,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4904,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4827,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8308634020618557,
+      "eval_loss": 0.4661047160625458,
+      "eval_runtime": 122.592,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4411,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4854,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5067,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4561,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3969,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4659,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4459,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4026,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4394,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.415,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8392396907216495,
+      "eval_loss": 0.44272318482398987,
+      "eval_runtime": 122.5862,
+      "eval_samples_per_second": 25.321,
+      "eval_steps_per_second": 3.165,
+      "step": 1750
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 7362045626142720.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.8291576786538372e-06,
+    "per_device_eval_batch_size": 8,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-9/checkpoint-1750/training_args.bin b/run-9/checkpoint-1750/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d01748e8cc129c4b09069f528a297f98c941f37e
--- /dev/null
+++ b/run-9/checkpoint-1750/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70462f89ef937a1a86b5f8cab0925039614416b5adeac1118b8269d0bb3e4722
+size 4728
diff --git a/run-9/checkpoint-2000/config.json b/run-9/checkpoint-2000/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-9/checkpoint-2000/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-9/checkpoint-2000/model.safetensors b/run-9/checkpoint-2000/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5edcd4b3708dad030076e0756cc62c3bac624518
--- /dev/null
+++ b/run-9/checkpoint-2000/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61ade75a9e1b1ecceaf9c3b6d4d2b432f7942903a2c8ff777c087a102f597af8
+size 433279996
diff --git a/run-9/checkpoint-2000/optimizer.pt b/run-9/checkpoint-2000/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..75fe8f26da52789f4469cc887d09674410fbf54d
--- /dev/null
+++ b/run-9/checkpoint-2000/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abf550de505e93236aafdea16088c2fc9d9641e0710fb9a6dcfbac52a953c20b
+size 866681082
diff --git a/run-9/checkpoint-2000/rng_state.pth b/run-9/checkpoint-2000/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..46bfd402f6bfeab3339ef84c13f8bea7848fb6b2
--- /dev/null
+++ b/run-9/checkpoint-2000/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5e36c9a58997a8fd5116115315a1541147a866ccc21b5fd8f202de7af5118db
+size 14244
diff --git a/run-9/checkpoint-2000/scheduler.pt b/run-9/checkpoint-2000/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..eceb3815b3745f2bef9329ebf24523948ce8f598
--- /dev/null
+++ b/run-9/checkpoint-2000/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:097a5497f7b984c5ac3a57126cc83e212503929cd40590ee21ece1f9e741389b
+size 1064
diff --git a/run-9/checkpoint-2000/trainer_state.json b/run-9/checkpoint-2000/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..778177374daf437b693f78b6aa7fc540595264ca
--- /dev/null
+++ b/run-9/checkpoint-2000/trainer_state.json
@@ -0,0 +1,576 @@
+{
+  "best_metric": 0.8456829896907216,
+  "best_model_checkpoint": "bert_trainer/run-9/checkpoint-2000",
+  "epoch": 2.5773195876288657,
+  "eval_steps": 250,
+  "global_step": 2000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.145788393269186e-07,
+      "loss": 1.4915,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.3598,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.236,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.1509,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0855,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0437,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0172,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9849,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9064,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9487,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.6591494845360825,
+      "eval_loss": 0.8960067629814148,
+      "eval_runtime": 122.5176,
+      "eval_samples_per_second": 25.335,
+      "eval_steps_per_second": 3.167,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8996,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9016,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8801,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8536,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7811,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7948,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7839,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7311,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7656,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7055,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.7625644329896907,
+      "eval_loss": 0.6912322640419006,
+      "eval_runtime": 122.6001,
+      "eval_samples_per_second": 25.318,
+      "eval_steps_per_second": 3.165,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7169,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6942,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6771,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6848,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6452,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5975,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6137,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6439,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.583,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5757,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.7976804123711341,
+      "eval_loss": 0.6034093499183655,
+      "eval_runtime": 122.5716,
+      "eval_samples_per_second": 25.324,
+      "eval_steps_per_second": 3.165,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5678,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5788,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6069,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5188,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.541,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5249,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5842,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5603,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5493,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5698,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8115335051546392,
+      "eval_loss": 0.5344901084899902,
+      "eval_runtime": 122.5922,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5672,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5462,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5508,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4896,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4862,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4929,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5232,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5364,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5632,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5132,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8273195876288659,
+      "eval_loss": 0.48882216215133667,
+      "eval_runtime": 122.5903,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.576,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4969,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4696,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4845,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4698,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4927,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5179,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4551,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4904,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4827,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8308634020618557,
+      "eval_loss": 0.4661047160625458,
+      "eval_runtime": 122.592,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4411,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4854,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5067,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4561,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3969,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4659,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4459,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4026,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4394,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.415,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8392396907216495,
+      "eval_loss": 0.44272318482398987,
+      "eval_runtime": 122.5862,
+      "eval_samples_per_second": 25.321,
+      "eval_steps_per_second": 3.165,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4331,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3842,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3963,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3876,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3826,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4407,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4261,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.394,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4411,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3936,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8456829896907216,
+      "eval_loss": 0.43213754892349243,
+      "eval_runtime": 122.5974,
+      "eval_samples_per_second": 25.319,
+      "eval_steps_per_second": 3.165,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 8414518195998720.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.8291576786538372e-06,
+    "per_device_eval_batch_size": 8,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-9/checkpoint-2000/training_args.bin b/run-9/checkpoint-2000/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d01748e8cc129c4b09069f528a297f98c941f37e
--- /dev/null
+++ b/run-9/checkpoint-2000/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70462f89ef937a1a86b5f8cab0925039614416b5adeac1118b8269d0bb3e4722
+size 4728
diff --git a/run-9/checkpoint-2250/config.json b/run-9/checkpoint-2250/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c3899aaf4604cb81deca05d01babb8d08484ab6
--- /dev/null
+++ b/run-9/checkpoint-2250/config.json
@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "bert-base-cased",
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.36.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}
diff --git a/run-9/checkpoint-2250/model.safetensors b/run-9/checkpoint-2250/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3e8a129f2f0fd32b0b233d578a79df2870cfe2e5
--- /dev/null
+++ b/run-9/checkpoint-2250/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb46a77a57c38c8c96b5a287b1dffc3d50221a70734933d6f7694c8491803332
+size 433279996
diff --git a/run-9/checkpoint-2250/optimizer.pt b/run-9/checkpoint-2250/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e27f641dbddf8d26518ab482d166445f73c68882
--- /dev/null
+++ b/run-9/checkpoint-2250/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ae503be0b0058dcf49c05d5d025b31e43c09f4e36203012a122a6fed87c23d8
+size 866681082
diff --git a/run-9/checkpoint-2250/rng_state.pth b/run-9/checkpoint-2250/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d7642df94958e9a26c811832ab6a75986dfa6876
--- /dev/null
+++ b/run-9/checkpoint-2250/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f90add1f686600af2c1d5c6844b41f65af21e9e1beb1e004c45df7efb0892f3
+size 14244
diff --git a/run-9/checkpoint-2250/scheduler.pt b/run-9/checkpoint-2250/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0e48b8af1d86d47e88e026a6caf55653e4a42da2
--- /dev/null
+++ b/run-9/checkpoint-2250/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bb98d4bf3e7fcd5cb78f5063ca1871a7884fd8be89f2162e95b6f04a78ddba9
+size 1064
diff --git a/run-9/checkpoint-2250/trainer_state.json b/run-9/checkpoint-2250/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..5b442b3a2c88bc3fd10d77b0fed695eee7e5e494
--- /dev/null
+++ b/run-9/checkpoint-2250/trainer_state.json
@@ -0,0 +1,645 @@
+{
+  "best_metric": 0.8505154639175257,
+  "best_model_checkpoint": "bert_trainer/run-9/checkpoint-2250",
+  "epoch": 2.899484536082474,
+  "eval_steps": 250,
+  "global_step": 2250,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.145788393269186e-07,
+      "loss": 1.4915,
+      "step": 25
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.3598,
+      "step": 50
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.236,
+      "step": 75
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.1509,
+      "step": 100
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0855,
+      "step": 125
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0437,
+      "step": 150
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 1.0172,
+      "step": 175
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9849,
+      "step": 200
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9064,
+      "step": 225
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9487,
+      "step": 250
+    },
+    {
+      "epoch": 0.32,
+      "eval_accuracy": 0.6591494845360825,
+      "eval_loss": 0.8960067629814148,
+      "eval_runtime": 122.5176,
+      "eval_samples_per_second": 25.335,
+      "eval_steps_per_second": 3.167,
+      "step": 250
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8996,
+      "step": 275
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.9016,
+      "step": 300
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8801,
+      "step": 325
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.8536,
+      "step": 350
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7811,
+      "step": 375
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7948,
+      "step": 400
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7839,
+      "step": 425
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7311,
+      "step": 450
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7656,
+      "step": 475
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7055,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "eval_accuracy": 0.7625644329896907,
+      "eval_loss": 0.6912322640419006,
+      "eval_runtime": 122.6001,
+      "eval_samples_per_second": 25.318,
+      "eval_steps_per_second": 3.165,
+      "step": 500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.7169,
+      "step": 525
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6942,
+      "step": 550
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6771,
+      "step": 575
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6848,
+      "step": 600
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6452,
+      "step": 625
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5975,
+      "step": 650
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6137,
+      "step": 675
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6439,
+      "step": 700
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.583,
+      "step": 725
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5757,
+      "step": 750
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.7976804123711341,
+      "eval_loss": 0.6034093499183655,
+      "eval_runtime": 122.5716,
+      "eval_samples_per_second": 25.324,
+      "eval_steps_per_second": 3.165,
+      "step": 750
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5678,
+      "step": 775
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5788,
+      "step": 800
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.6069,
+      "step": 825
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5188,
+      "step": 850
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.541,
+      "step": 875
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5249,
+      "step": 900
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5842,
+      "step": 925
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5603,
+      "step": 950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5493,
+      "step": 975
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5698,
+      "step": 1000
+    },
+    {
+      "epoch": 1.29,
+      "eval_accuracy": 0.8115335051546392,
+      "eval_loss": 0.5344901084899902,
+      "eval_runtime": 122.5922,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5672,
+      "step": 1025
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5462,
+      "step": 1050
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5508,
+      "step": 1075
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4896,
+      "step": 1100
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4862,
+      "step": 1125
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4929,
+      "step": 1150
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5232,
+      "step": 1175
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5364,
+      "step": 1200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5632,
+      "step": 1225
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5132,
+      "step": 1250
+    },
+    {
+      "epoch": 1.61,
+      "eval_accuracy": 0.8273195876288659,
+      "eval_loss": 0.48882216215133667,
+      "eval_runtime": 122.5903,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1250
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.576,
+      "step": 1275
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4969,
+      "step": 1300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4696,
+      "step": 1325
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4845,
+      "step": 1350
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4698,
+      "step": 1375
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4927,
+      "step": 1400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5179,
+      "step": 1425
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4551,
+      "step": 1450
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4904,
+      "step": 1475
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4827,
+      "step": 1500
+    },
+    {
+      "epoch": 1.93,
+      "eval_accuracy": 0.8308634020618557,
+      "eval_loss": 0.4661047160625458,
+      "eval_runtime": 122.592,
+      "eval_samples_per_second": 25.32,
+      "eval_steps_per_second": 3.165,
+      "step": 1500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4411,
+      "step": 1525
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4854,
+      "step": 1550
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.5067,
+      "step": 1575
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4561,
+      "step": 1600
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3969,
+      "step": 1625
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4659,
+      "step": 1650
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4459,
+      "step": 1675
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4026,
+      "step": 1700
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4394,
+      "step": 1725
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.415,
+      "step": 1750
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 0.8392396907216495,
+      "eval_loss": 0.44272318482398987,
+      "eval_runtime": 122.5862,
+      "eval_samples_per_second": 25.321,
+      "eval_steps_per_second": 3.165,
+      "step": 1750
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4331,
+      "step": 1775
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3842,
+      "step": 1800
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3963,
+      "step": 1825
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3876,
+      "step": 1850
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3826,
+      "step": 1875
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4407,
+      "step": 1900
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4261,
+      "step": 1925
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.394,
+      "step": 1950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4411,
+      "step": 1975
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3936,
+      "step": 2000
+    },
+    {
+      "epoch": 2.58,
+      "eval_accuracy": 0.8456829896907216,
+      "eval_loss": 0.43213754892349243,
+      "eval_runtime": 122.5974,
+      "eval_samples_per_second": 25.319,
+      "eval_steps_per_second": 3.165,
+      "step": 2000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4434,
+      "step": 2025
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4225,
+      "step": 2050
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3864,
+      "step": 2075
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4286,
+      "step": 2100
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4143,
+      "step": 2125
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3637,
+      "step": 2150
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4172,
+      "step": 2175
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3794,
+      "step": 2200
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.3987,
+      "step": 2225
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 1.8291576786538372e-06,
+      "loss": 0.4932,
+      "step": 2250
+    },
+    {
+      "epoch": 2.9,
+      "eval_accuracy": 0.8505154639175257,
+      "eval_loss": 0.42465415596961975,
+      "eval_runtime": 122.5883,
+      "eval_samples_per_second": 25.321,
+      "eval_steps_per_second": 3.165,
+      "step": 2250
+    }
+  ],
+  "logging_steps": 25,
+  "max_steps": 3000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 250,
+  "total_flos": 9466990765854720.0,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.8291576786538372e-06,
+    "per_device_eval_batch_size": 8,
+    "per_device_train_batch_size": 16
+  }
+}
diff --git a/run-9/checkpoint-2250/training_args.bin b/run-9/checkpoint-2250/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..d01748e8cc129c4b09069f528a297f98c941f37e
--- /dev/null
+++ b/run-9/checkpoint-2250/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70462f89ef937a1a86b5f8cab0925039614416b5adeac1118b8269d0bb3e4722
+size 4728
diff --git a/runs/Dec25_09-34-21_ArtanisPC/events.out.tfevents.1703493273.ArtanisPC.507.14 b/runs/Dec25_09-34-21_ArtanisPC/events.out.tfevents.1703493273.ArtanisPC.507.14
new file mode 100644
index 0000000000000000000000000000000000000000..d06cfcb025e7b9dee841f30f27af5590ae052c4b
--- /dev/null
+++ b/runs/Dec25_09-34-21_ArtanisPC/events.out.tfevents.1703493273.ArtanisPC.507.14
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa863999cbf0a0692ac89f72cdfba75c55aaf4dfa3bc8795378fa271f5eb23a5
+size 6060
diff --git a/runs/Dec25_09-37-20_ArtanisPC/events.out.tfevents.1703493442.ArtanisPC.507.15 b/runs/Dec25_09-37-20_ArtanisPC/events.out.tfevents.1703493442.ArtanisPC.507.15
new file mode 100644
index 0000000000000000000000000000000000000000..38f7ac5b1394e6101e58e0f089f61c8f198ee0e8
--- /dev/null
+++ b/runs/Dec25_09-37-20_ArtanisPC/events.out.tfevents.1703493442.ArtanisPC.507.15
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6365e47e7a964d1ba040201cad21bfb29f63d3d7e7005d8973cd149518ad371
+size 4489
diff --git a/runs/Dec25_09-37-31_ArtanisPC/events.out.tfevents.1703493459.ArtanisPC.507.16 b/runs/Dec25_09-37-31_ArtanisPC/events.out.tfevents.1703493459.ArtanisPC.507.16
new file mode 100644
index 0000000000000000000000000000000000000000..ca3331cacb2028dd2f904d67338a05f4fe7891db
--- /dev/null
+++ b/runs/Dec25_09-37-31_ArtanisPC/events.out.tfevents.1703493459.ArtanisPC.507.16
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6419aa6f7aed4852e6fc6918966178a73e33bf32a727286f9067b58f8f0d1f15
+size 7953
diff --git a/runs/Dec25_09-39-44_ArtanisPC/events.out.tfevents.1703493592.ArtanisPC.507.17 b/runs/Dec25_09-39-44_ArtanisPC/events.out.tfevents.1703493592.ArtanisPC.507.17
new file mode 100644
index 0000000000000000000000000000000000000000..6d11e95f7ff44f0f05ca930ddbd6f61c20cd6896
--- /dev/null
+++ b/runs/Dec25_09-39-44_ArtanisPC/events.out.tfevents.1703493592.ArtanisPC.507.17
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1e870f4da326da768f87c93442dcd25392d5304a66b5e50c3d2ab20fe785c85
+size 7324
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703496003.ArtanisPC.507.18 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703496003.ArtanisPC.507.18
new file mode 100644
index 0000000000000000000000000000000000000000..89d9000addf5c086c625787c447ff4128bff468e
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703496003.ArtanisPC.507.18
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be4647903afabf224835280a16b928b5a049a89429755bffdaaf0e705d26e610
+size 27558
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703497105.ArtanisPC.507.19 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703497105.ArtanisPC.507.19
new file mode 100644
index 0000000000000000000000000000000000000000..dd436176a207b232c25136d428b768fdf336e845
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703497105.ArtanisPC.507.19
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1db979601960139b7005d2a945e6ba8c3f96fcc0751d228e50dd470a478ebc6
+size 27558
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703498188.ArtanisPC.507.20 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703498188.ArtanisPC.507.20
new file mode 100644
index 0000000000000000000000000000000000000000..53b4aef2fd31536dee8eb2e0dbb3eced302830a0
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703498188.ArtanisPC.507.20
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40b69a51b8faf8632ee16f32c043a206a5a10e2f2d4d4ee6bd48b94caeb9c970
+size 27559
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703498954.ArtanisPC.507.21 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703498954.ArtanisPC.507.21
new file mode 100644
index 0000000000000000000000000000000000000000..2c015fe13736f6378daa0149b642d4deaee71351
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703498954.ArtanisPC.507.21
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:009658267dac2c96b000746bc4c3ddd40a1a27d5ce4323c9d25b5a3c8b4972d9
+size 27559
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703499720.ArtanisPC.507.22 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703499720.ArtanisPC.507.22
new file mode 100644
index 0000000000000000000000000000000000000000..9a647b90c4883289b9d919eccb3b65df096a77bd
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703499720.ArtanisPC.507.22
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa2310aeefecb3999998b0482c67d230a6625976f7aeed1f458233d6e95ee43f
+size 27558
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703500799.ArtanisPC.507.23 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703500799.ArtanisPC.507.23
new file mode 100644
index 0000000000000000000000000000000000000000..e8444c02fee234fa533761f56c67735d31669fd6
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703500799.ArtanisPC.507.23
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b8483e756dd2b0015d2af86ee405474b7b06659848f2a6de0e07c8ca5c51e72
+size 27558
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703501563.ArtanisPC.507.24 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703501563.ArtanisPC.507.24
new file mode 100644
index 0000000000000000000000000000000000000000..da58afddfc4957c07fe317c39603985cbb6c10af
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703501563.ArtanisPC.507.24
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8acc946abdd3994c2c1c1eaccbd211688ed1a375ea91798c017621b63a37b479
+size 27561
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703509610.ArtanisPC.507.25 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703509610.ArtanisPC.507.25
new file mode 100644
index 0000000000000000000000000000000000000000..05e4bc5b3623b987825a0af5d060193396ebc4ec
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703509610.ArtanisPC.507.25
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:070db139e49a1c5f661de241bd31e6e453941c1ad95996faec0b3ccdd09f7530
+size 6381
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703510367.ArtanisPC.507.26 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703510367.ArtanisPC.507.26
new file mode 100644
index 0000000000000000000000000000000000000000..42598eb0176e183ef0979313f906c3ea70621167
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703510367.ArtanisPC.507.26
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b31d6f0ac4a77c40db807537eae3546f7d4b44f1813169447633e0806c0bafa
+size 6382
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703510459.ArtanisPC.507.27 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703510459.ArtanisPC.507.27
new file mode 100644
index 0000000000000000000000000000000000000000..f4623af3eb463ba8358b1092e96e0d4ea4d69eba
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703510459.ArtanisPC.507.27
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1305693f59568093043807faacf79763407d39cb960e6ededde8e240f8496768
+size 49203
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703526087.ArtanisPC.507.28 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703526087.ArtanisPC.507.28
new file mode 100644
index 0000000000000000000000000000000000000000..c3d8085dc42e0df94c212401722c5171155ceec1
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703526087.ArtanisPC.507.28
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e58b4ecbae955485a0c5bf21f482849d884b5427202e8ebd6df3df44eef1a49
+size 27560
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703534696.ArtanisPC.507.29 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703534696.ArtanisPC.507.29
new file mode 100644
index 0000000000000000000000000000000000000000..a0e674a31c54621a3b3a45b3a2045571db853fe7
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703534696.ArtanisPC.507.29
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7f2b8e60071bf6b5b6233a3596ec53ec6b57408791ed086206dfd51e611e1b8
+size 27559
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703535781.ArtanisPC.507.30 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703535781.ArtanisPC.507.30
new file mode 100644
index 0000000000000000000000000000000000000000..b96a83496b4928d55361fbbf70f4b43464141b26
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703535781.ArtanisPC.507.30
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0800a0bfcf9d9ab09485a7e9ffc17e871017eaa43c0821e4cdddb7b58607f8b3
+size 27560
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703536867.ArtanisPC.507.31 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703536867.ArtanisPC.507.31
new file mode 100644
index 0000000000000000000000000000000000000000..7a39829cd001ffbc01bc1b008ebdeebd6a514955
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703536867.ArtanisPC.507.31
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef1965518ef1d530291d877aa75bb4b48aeb7aaac04ee219d2c9f53ef00b88ca
+size 27560
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703545693.ArtanisPC.507.32 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703545693.ArtanisPC.507.32
new file mode 100644
index 0000000000000000000000000000000000000000..a52d9bbe0a0001f3fc9aba906f5134303f4bf927
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703545693.ArtanisPC.507.32
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c562058a31a0bc63268bc9513f70d3f008c21454e0269dc3a86969cf71aa705a
+size 15847
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549538.ArtanisPC.507.33 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549538.ArtanisPC.507.33
new file mode 100644
index 0000000000000000000000000000000000000000..bbc4b9a4f96c94ef3c239cf33be7d466eb0c2a1d
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549538.ArtanisPC.507.33
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70d4cddd7b37481768c916fc2b70316a5659b8f5cae21370f7a336476ee89eab
+size 6381
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549629.ArtanisPC.507.34 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549629.ArtanisPC.507.34
new file mode 100644
index 0000000000000000000000000000000000000000..a523a555dfaccfc1eac7dba7ad9620afdbd422b1
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549629.ArtanisPC.507.34
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2f105231b83f58ab6a0201c269b0488d70e7cf3534facbd5bbae5520377d87a
+size 6381
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549718.ArtanisPC.507.35 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549718.ArtanisPC.507.35
new file mode 100644
index 0000000000000000000000000000000000000000..d4850b0f0a8ff597e8724e0bcbfaaa12999c9716
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549718.ArtanisPC.507.35
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ee2fa3ea1ef4031237daed8530c86c6432a60472cabe747d1d7e48735b21a6f
+size 6381
diff --git a/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549781.ArtanisPC.507.36 b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549781.ArtanisPC.507.36
new file mode 100644
index 0000000000000000000000000000000000000000..62492b261391889bb2a751cb3ac77ca5a9978e5e
--- /dev/null
+++ b/runs/Dec25_10-19-57_ArtanisPC/events.out.tfevents.1703549781.ArtanisPC.507.36
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42726df16ef685e792103fbca9d3131234fe35f4b1fb474e3e6bd7e7ee71ef1d
+size 6381
diff --git a/training_args.bin b/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..3e199042dd22f2119217ad3739574886d8683746
--- /dev/null
+++ b/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6ba6115481ac0d03baaa980ba32ca8fce3208b8d5ad2bcc99c60e691a88223f
+size 4728