Training in progress, epoch 1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:496a7911d6025b5f84fac88f6ccd974edd5c92e3f0c11cba8d02debb78aacef1
+oid sha256:97687f5fed50ebc5579f2887c4a528f3caa14d8822166df6c145698780fc26c1
 size 438100144

run-3/checkpoint-11995/config.json

@@ -0,0 +1,125 @@
+{
+  "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",
+    "5": "LABEL_5",
+    "6": "LABEL_6",
+    "7": "LABEL_7",
+    "8": "LABEL_8",
+    "9": "LABEL_9",
+    "10": "LABEL_10",
+    "11": "LABEL_11",
+    "12": "LABEL_12",
+    "13": "LABEL_13",
+    "14": "LABEL_14",
+    "15": "LABEL_15",
+    "16": "LABEL_16",
+    "17": "LABEL_17",
+    "18": "LABEL_18",
+    "19": "LABEL_19",
+    "20": "LABEL_20",
+    "21": "LABEL_21",
+    "22": "LABEL_22",
+    "23": "LABEL_23",
+    "24": "LABEL_24",
+    "25": "LABEL_25",
+    "26": "LABEL_26",
+    "27": "LABEL_27",
+    "28": "LABEL_28",
+    "29": "LABEL_29",
+    "30": "LABEL_30",
+    "31": "LABEL_31",
+    "32": "LABEL_32",
+    "33": "LABEL_33",
+    "34": "LABEL_34",
+    "35": "LABEL_35",
+    "36": "LABEL_36",
+    "37": "LABEL_37",
+    "38": "LABEL_38",
+    "39": "LABEL_39",
+    "40": "LABEL_40",
+    "41": "LABEL_41",
+    "42": "LABEL_42",
+    "43": "LABEL_43",
+    "44": "LABEL_44",
+    "45": "LABEL_45",
+    "46": "LABEL_46",
+    "47": "LABEL_47"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_10": 10,
+    "LABEL_11": 11,
+    "LABEL_12": 12,
+    "LABEL_13": 13,
+    "LABEL_14": 14,
+    "LABEL_15": 15,
+    "LABEL_16": 16,
+    "LABEL_17": 17,
+    "LABEL_18": 18,
+    "LABEL_19": 19,
+    "LABEL_2": 2,
+    "LABEL_20": 20,
+    "LABEL_21": 21,
+    "LABEL_22": 22,
+    "LABEL_23": 23,
+    "LABEL_24": 24,
+    "LABEL_25": 25,
+    "LABEL_26": 26,
+    "LABEL_27": 27,
+    "LABEL_28": 28,
+    "LABEL_29": 29,
+    "LABEL_3": 3,
+    "LABEL_30": 30,
+    "LABEL_31": 31,
+    "LABEL_32": 32,
+    "LABEL_33": 33,
+    "LABEL_34": 34,
+    "LABEL_35": 35,
+    "LABEL_36": 36,
+    "LABEL_37": 37,
+    "LABEL_38": 38,
+    "LABEL_39": 39,
+    "LABEL_4": 4,
+    "LABEL_40": 40,
+    "LABEL_41": 41,
+    "LABEL_42": 42,
+    "LABEL_43": 43,
+    "LABEL_44": 44,
+    "LABEL_45": 45,
+    "LABEL_46": 46,
+    "LABEL_47": 47,
+    "LABEL_5": 5,
+    "LABEL_6": 6,
+    "LABEL_7": 7,
+    "LABEL_8": 8,
+    "LABEL_9": 9
+  },
+  "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",
+  "torch_dtype": "float32",
+  "transformers_version": "4.50.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-3/checkpoint-11995/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d9ca6a4726ae103ac8d05b8cd7032c222253ffa169222515410201fd20edca5
+size 438100144

run-3/checkpoint-11995/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:775df1e19ddc4c8b36068c6d66c38b6b2be4a1de4b1a702bb95616b612535b8a
+size 876321402

run-3/checkpoint-11995/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b3f51a81800533b5f405ae674a60a515fe53b8ffe03db1840c5ab046ef1725d
+size 14244

run-3/checkpoint-11995/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69e3ec9dad272af1e89055cef6280046ea66fe279e02bf79c2a31ff06895da39
+size 988

run-3/checkpoint-11995/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:949f596ed91c4eca380973999aef3661ae81ddd8ac4fcf3b2b90f76db90fab98
+size 1064

run-3/checkpoint-11995/trainer_state.json

@@ -0,0 +1,1050 @@
+{
+  "best_global_step": 7197,
+  "best_metric": 0.9341971854179941,
+  "best_model_checkpoint": "./results/run-3/checkpoint-7197",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 11995,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.041684035014589414,
+      "grad_norm": 1.7545671463012695,
+      "learning_rate": 1.386e-05,
+      "loss": 1.2954,
+      "step": 100
+    },
+    {
+      "epoch": 0.08336807002917883,
+      "grad_norm": 0.7225435376167297,
+      "learning_rate": 2.7859999999999998e-05,
+      "loss": 0.5134,
+      "step": 200
+    },
+    {
+      "epoch": 0.12505210504376824,
+      "grad_norm": 0.9463444948196411,
+      "learning_rate": 4.1859999999999996e-05,
+      "loss": 0.4399,
+      "step": 300
+    },
+    {
+      "epoch": 0.16673614005835766,
+      "grad_norm": 0.8181287050247192,
+      "learning_rate": 5.586e-05,
+      "loss": 0.3608,
+      "step": 400
+    },
+    {
+      "epoch": 0.20842017507294705,
+      "grad_norm": 1.4074561595916748,
+      "learning_rate": 6.986e-05,
+      "loss": 0.3303,
+      "step": 500
+    },
+    {
+      "epoch": 0.25010421008753647,
+      "grad_norm": 1.232994556427002,
+      "learning_rate": 6.98170731707317e-05,
+      "loss": 0.3118,
+      "step": 600
+    },
+    {
+      "epoch": 0.29178824510212586,
+      "grad_norm": 1.1151797771453857,
+      "learning_rate": 6.96341463414634e-05,
+      "loss": 0.3067,
+      "step": 700
+    },
+    {
+      "epoch": 0.3334722801167153,
+      "grad_norm": 1.3453563451766968,
+      "learning_rate": 6.944937176644493e-05,
+      "loss": 0.2959,
+      "step": 800
+    },
+    {
+      "epoch": 0.3751563151313047,
+      "grad_norm": 1.3014110326766968,
+      "learning_rate": 6.926459719142646e-05,
+      "loss": 0.2803,
+      "step": 900
+    },
+    {
+      "epoch": 0.4168403501458941,
+      "grad_norm": 2.4783546924591064,
+      "learning_rate": 6.907982261640798e-05,
+      "loss": 0.275,
+      "step": 1000
+    },
+    {
+      "epoch": 0.45852438516048355,
+      "grad_norm": 1.0991694927215576,
+      "learning_rate": 6.88950480413895e-05,
+      "loss": 0.2725,
+      "step": 1100
+    },
+    {
+      "epoch": 0.5002084201750729,
+      "grad_norm": 1.8534690141677856,
+      "learning_rate": 6.871027346637102e-05,
+      "loss": 0.2965,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5418924551896623,
+      "grad_norm": 1.533751130104065,
+      "learning_rate": 6.852549889135254e-05,
+      "loss": 0.257,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5835764902042517,
+      "grad_norm": 1.39555025100708,
+      "learning_rate": 6.834072431633407e-05,
+      "loss": 0.2666,
+      "step": 1400
+    },
+    {
+      "epoch": 0.6252605252188412,
+      "grad_norm": 1.1627668142318726,
+      "learning_rate": 6.81559497413156e-05,
+      "loss": 0.2461,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6669445602334306,
+      "grad_norm": 1.319166660308838,
+      "learning_rate": 6.79711751662971e-05,
+      "loss": 0.242,
+      "step": 1600
+    },
+    {
+      "epoch": 0.70862859524802,
+      "grad_norm": 1.3535823822021484,
+      "learning_rate": 6.778640059127863e-05,
+      "loss": 0.2463,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7503126302626094,
+      "grad_norm": 1.8609535694122314,
+      "learning_rate": 6.760162601626016e-05,
+      "loss": 0.2367,
+      "step": 1800
+    },
+    {
+      "epoch": 0.7919966652771988,
+      "grad_norm": 1.86128830909729,
+      "learning_rate": 6.741685144124168e-05,
+      "loss": 0.2449,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8336807002917882,
+      "grad_norm": 1.8068938255310059,
+      "learning_rate": 6.723207686622321e-05,
+      "loss": 0.2329,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8753647353063777,
+      "grad_norm": 1.4341034889221191,
+      "learning_rate": 6.704730229120472e-05,
+      "loss": 0.2342,
+      "step": 2100
+    },
+    {
+      "epoch": 0.9170487703209671,
+      "grad_norm": 0.7307045459747314,
+      "learning_rate": 6.686252771618624e-05,
+      "loss": 0.2358,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9587328053355565,
+      "grad_norm": 2.0308947563171387,
+      "learning_rate": 6.667775314116777e-05,
+      "loss": 0.2244,
+      "step": 2300
+    },
+    {
+      "epoch": 1.0,
+      "eval_administration_accuracy": 0.9317281634354805,
+      "eval_administration_f1": 0.9110301299328539,
+      "eval_corruption_accuracy": 0.9541380029184907,
+      "eval_corruption_f1": 0.9489574147600867,
+      "eval_democracy_accuracy": 0.9473629351678132,
+      "eval_democracy_f1": 0.935646230598331,
+      "eval_development_accuracy": 0.910464873879508,
+      "eval_development_f1": 0.9008646598644421,
+      "eval_economy_accuracy": 0.9235980821346675,
+      "eval_economy_f1": 0.9156600494130359,
+      "eval_education_accuracy": 0.9622680842193038,
+      "eval_education_f1": 0.9586233907920864,
+      "eval_environment_accuracy": 0.978319783197832,
+      "eval_environment_f1": 0.9770754665605412,
+      "eval_instability_accuracy": 0.9398582447362935,
+      "eval_instability_f1": 0.9299109706262821,
+      "eval_leadership_accuracy": 0.8081092349385033,
+      "eval_leadership_f1": 0.7818188134942813,
+      "eval_loss": 0.2264394313097,
+      "eval_overall_accuracy": 0.9310245987075256,
+      "eval_overall_f1": 0.9219705720705208,
+      "eval_race_accuracy": 0.9487179487179487,
+      "eval_race_f1": 0.9448562040025291,
+      "eval_religion_accuracy": 0.9527829893683553,
+      "eval_religion_f1": 0.950851997767183,
+      "eval_runtime": 11.1066,
+      "eval_safety_accuracy": 0.9149468417761101,
+      "eval_safety_f1": 0.9083515370345979,
+      "eval_samples_per_second": 863.809,
+      "eval_steps_per_second": 54.022,
+      "step": 2399
+    },
+    {
+      "epoch": 1.0004168403501459,
+      "grad_norm": 1.3034080266952515,
+      "learning_rate": 6.64929785661493e-05,
+      "loss": 0.2417,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0421008753647354,
+      "grad_norm": 1.3565603494644165,
+      "learning_rate": 6.630820399113082e-05,
+      "loss": 0.1946,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0837849103793247,
+      "grad_norm": 2.098259449005127,
+      "learning_rate": 6.612342941611233e-05,
+      "loss": 0.1911,
+      "step": 2600
+    },
+    {
+      "epoch": 1.1254689453939142,
+      "grad_norm": 1.0764553546905518,
+      "learning_rate": 6.593865484109386e-05,
+      "loss": 0.1821,
+      "step": 2700
+    },
+    {
+      "epoch": 1.1671529804085035,
+      "grad_norm": 1.4594541788101196,
+      "learning_rate": 6.575388026607538e-05,
+      "loss": 0.1851,
+      "step": 2800
+    },
+    {
+      "epoch": 1.208837015423093,
+      "grad_norm": 1.8065540790557861,
+      "learning_rate": 6.556910569105691e-05,
+      "loss": 0.1929,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2505210504376825,
+      "grad_norm": 0.8411226272583008,
+      "learning_rate": 6.538433111603843e-05,
+      "loss": 0.1882,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2922050854522718,
+      "grad_norm": 1.886057734489441,
+      "learning_rate": 6.519955654101994e-05,
+      "loss": 0.1861,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3338891204668613,
+      "grad_norm": 1.2633222341537476,
+      "learning_rate": 6.501478196600147e-05,
+      "loss": 0.1816,
+      "step": 3200
+    },
+    {
+      "epoch": 1.3755731554814505,
+      "grad_norm": 2.1033339500427246,
+      "learning_rate": 6.4830007390983e-05,
+      "loss": 0.1887,
+      "step": 3300
+    },
+    {
+      "epoch": 1.41725719049604,
+      "grad_norm": 1.5505166053771973,
+      "learning_rate": 6.464523281596452e-05,
+      "loss": 0.1946,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4589412255106295,
+      "grad_norm": 1.7376617193222046,
+      "learning_rate": 6.446045824094603e-05,
+      "loss": 0.1881,
+      "step": 3500
+    },
+    {
+      "epoch": 1.5006252605252188,
+      "grad_norm": 1.8713116645812988,
+      "learning_rate": 6.427568366592756e-05,
+      "loss": 0.1742,
+      "step": 3600
+    },
+    {
+      "epoch": 1.5423092955398081,
+      "grad_norm": 1.2346725463867188,
+      "learning_rate": 6.409090909090908e-05,
+      "loss": 0.1733,
+      "step": 3700
+    },
+    {
+      "epoch": 1.5839933305543976,
+      "grad_norm": 1.5446250438690186,
+      "learning_rate": 6.390613451589061e-05,
+      "loss": 0.1885,
+      "step": 3800
+    },
+    {
+      "epoch": 1.6256773655689871,
+      "grad_norm": 2.538613796234131,
+      "learning_rate": 6.372135994087213e-05,
+      "loss": 0.1737,
+      "step": 3900
+    },
+    {
+      "epoch": 1.6673614005835766,
+      "grad_norm": 1.1538047790527344,
+      "learning_rate": 6.353658536585365e-05,
+      "loss": 0.1836,
+      "step": 4000
+    },
+    {
+      "epoch": 1.709045435598166,
+      "grad_norm": 2.285994291305542,
+      "learning_rate": 6.335181079083517e-05,
+      "loss": 0.1934,
+      "step": 4100
+    },
+    {
+      "epoch": 1.7507294706127552,
+      "grad_norm": 1.3632721900939941,
+      "learning_rate": 6.31670362158167e-05,
+      "loss": 0.1792,
+      "step": 4200
+    },
+    {
+      "epoch": 1.7924135056273447,
+      "grad_norm": 1.3432073593139648,
+      "learning_rate": 6.298226164079822e-05,
+      "loss": 0.1665,
+      "step": 4300
+    },
+    {
+      "epoch": 1.8340975406419342,
+      "grad_norm": 4.086724758148193,
+      "learning_rate": 6.279748706577975e-05,
+      "loss": 0.1761,
+      "step": 4400
+    },
+    {
+      "epoch": 1.8757815756565237,
+      "grad_norm": 1.1902227401733398,
+      "learning_rate": 6.261271249076126e-05,
+      "loss": 0.1738,
+      "step": 4500
+    },
+    {
+      "epoch": 1.917465610671113,
+      "grad_norm": 2.1310391426086426,
+      "learning_rate": 6.242793791574278e-05,
+      "loss": 0.1822,
+      "step": 4600
+    },
+    {
+      "epoch": 1.9591496456857023,
+      "grad_norm": 1.384017825126648,
+      "learning_rate": 6.224316334072431e-05,
+      "loss": 0.1827,
+      "step": 4700
+    },
+    {
+      "epoch": 2.0,
+      "eval_administration_accuracy": 0.9382947675630603,
+      "eval_administration_f1": 0.93083241604641,
+      "eval_corruption_accuracy": 0.9589326662497394,
+      "eval_corruption_f1": 0.9564057119817679,
+      "eval_democracy_accuracy": 0.9522618303106108,
+      "eval_democracy_f1": 0.9459837872310805,
+      "eval_development_accuracy": 0.9119241192411924,
+      "eval_development_f1": 0.9077803592136469,
+      "eval_economy_accuracy": 0.9272461955388784,
+      "eval_economy_f1": 0.9256537091607892,
+      "eval_education_accuracy": 0.9613299979153638,
+      "eval_education_f1": 0.9608415429902122,
+      "eval_environment_accuracy": 0.978319783197832,
+      "eval_environment_f1": 0.9774790124713261,
+      "eval_instability_accuracy": 0.9247446320617052,
+      "eval_instability_f1": 0.9296973340071476,
+      "eval_leadership_accuracy": 0.8296852199291224,
+      "eval_leadership_f1": 0.8138352706726838,
+      "eval_loss": 0.21539363265037537,
+      "eval_overall_accuracy": 0.9336477659648392,
+      "eval_overall_f1": 0.9304371117477391,
+      "eval_race_accuracy": 0.9497602668334376,
+      "eval_race_f1": 0.9482550954258615,
+      "eval_religion_accuracy": 0.9559099437148217,
+      "eval_religion_f1": 0.9545949788562016,
+      "eval_runtime": 11.0609,
+      "eval_safety_accuracy": 0.9153637690223057,
+      "eval_safety_f1": 0.9138861229157411,
+      "eval_samples_per_second": 867.383,
+      "eval_steps_per_second": 54.245,
+      "step": 4798
+    },
+    {
+      "epoch": 2.0008336807002918,
+      "grad_norm": 0.8514364957809448,
+      "learning_rate": 6.205838876570583e-05,
+      "loss": 0.1822,
+      "step": 4800
+    },
+    {
+      "epoch": 2.0425177157148813,
+      "grad_norm": 1.2515586614608765,
+      "learning_rate": 6.187361419068736e-05,
+      "loss": 0.1312,
+      "step": 4900
+    },
+    {
+      "epoch": 2.084201750729471,
+      "grad_norm": 1.2924548387527466,
+      "learning_rate": 6.168883961566887e-05,
+      "loss": 0.1307,
+      "step": 5000
+    },
+    {
+      "epoch": 2.12588578574406,
+      "grad_norm": 1.7066774368286133,
+      "learning_rate": 6.15040650406504e-05,
+      "loss": 0.1248,
+      "step": 5100
+    },
+    {
+      "epoch": 2.1675698207586493,
+      "grad_norm": 1.4067167043685913,
+      "learning_rate": 6.131929046563192e-05,
+      "loss": 0.1318,
+      "step": 5200
+    },
+    {
+      "epoch": 2.209253855773239,
+      "grad_norm": 2.4648754596710205,
+      "learning_rate": 6.113451589061345e-05,
+      "loss": 0.1293,
+      "step": 5300
+    },
+    {
+      "epoch": 2.2509378907878284,
+      "grad_norm": 2.232621669769287,
+      "learning_rate": 6.0949741315594965e-05,
+      "loss": 0.128,
+      "step": 5400
+    },
+    {
+      "epoch": 2.292621925802418,
+      "grad_norm": 2.5116114616394043,
+      "learning_rate": 6.076496674057649e-05,
+      "loss": 0.1328,
+      "step": 5500
+    },
+    {
+      "epoch": 2.334305960817007,
+      "grad_norm": 1.2505563497543335,
+      "learning_rate": 6.0580192165558015e-05,
+      "loss": 0.1284,
+      "step": 5600
+    },
+    {
+      "epoch": 2.3759899958315964,
+      "grad_norm": 1.3521802425384521,
+      "learning_rate": 6.039541759053954e-05,
+      "loss": 0.1345,
+      "step": 5700
+    },
+    {
+      "epoch": 2.417674030846186,
+      "grad_norm": 1.0144108533859253,
+      "learning_rate": 6.021064301552106e-05,
+      "loss": 0.1366,
+      "step": 5800
+    },
+    {
+      "epoch": 2.4593580658607754,
+      "grad_norm": 0.8403565287590027,
+      "learning_rate": 6.002586844050258e-05,
+      "loss": 0.1397,
+      "step": 5900
+    },
+    {
+      "epoch": 2.501042100875365,
+      "grad_norm": 2.634355068206787,
+      "learning_rate": 5.98410938654841e-05,
+      "loss": 0.1373,
+      "step": 6000
+    },
+    {
+      "epoch": 2.542726135889954,
+      "grad_norm": 3.5822253227233887,
+      "learning_rate": 5.965631929046563e-05,
+      "loss": 0.1291,
+      "step": 6100
+    },
+    {
+      "epoch": 2.5844101709045435,
+      "grad_norm": 1.4991543292999268,
+      "learning_rate": 5.947154471544715e-05,
+      "loss": 0.1475,
+      "step": 6200
+    },
+    {
+      "epoch": 2.626094205919133,
+      "grad_norm": 1.9590423107147217,
+      "learning_rate": 5.928677014042867e-05,
+      "loss": 0.1411,
+      "step": 6300
+    },
+    {
+      "epoch": 2.6677782409337225,
+      "grad_norm": 1.2120991945266724,
+      "learning_rate": 5.910199556541019e-05,
+      "loss": 0.1397,
+      "step": 6400
+    },
+    {
+      "epoch": 2.7094622759483116,
+      "grad_norm": 2.270709991455078,
+      "learning_rate": 5.8917220990391715e-05,
+      "loss": 0.1398,
+      "step": 6500
+    },
+    {
+      "epoch": 2.751146310962901,
+      "grad_norm": 2.1711227893829346,
+      "learning_rate": 5.873244641537324e-05,
+      "loss": 0.1344,
+      "step": 6600
+    },
+    {
+      "epoch": 2.7928303459774906,
+      "grad_norm": 3.177982807159424,
+      "learning_rate": 5.854951958610495e-05,
+      "loss": 0.1377,
+      "step": 6700
+    },
+    {
+      "epoch": 2.83451438099208,
+      "grad_norm": 1.8457061052322388,
+      "learning_rate": 5.8364745011086465e-05,
+      "loss": 0.1277,
+      "step": 6800
+    },
+    {
+      "epoch": 2.8761984160066696,
+      "grad_norm": 1.9990291595458984,
+      "learning_rate": 5.817997043606799e-05,
+      "loss": 0.1311,
+      "step": 6900
+    },
+    {
+      "epoch": 2.917882451021259,
+      "grad_norm": 1.8258967399597168,
+      "learning_rate": 5.7995195861049516e-05,
+      "loss": 0.1359,
+      "step": 7000
+    },
+    {
+      "epoch": 2.959566486035848,
+      "grad_norm": 0.9961782097816467,
+      "learning_rate": 5.781042128603104e-05,
+      "loss": 0.1416,
+      "step": 7100
+    },
+    {
+      "epoch": 3.0,
+      "eval_administration_accuracy": 0.9374609130706691,
+      "eval_administration_f1": 0.9317222442324985,
+      "eval_corruption_accuracy": 0.9572649572649573,
+      "eval_corruption_f1": 0.9537446744289325,
+      "eval_democracy_accuracy": 0.9529914529914529,
+      "eval_democracy_f1": 0.9494686606905594,
+      "eval_development_accuracy": 0.9147383781530123,
+      "eval_development_f1": 0.9100235798889205,
+      "eval_economy_accuracy": 0.9329789451740671,
+      "eval_economy_f1": 0.9281892557992357,
+      "eval_education_accuracy": 0.964665415884928,
+      "eval_education_f1": 0.9620886841561839,
+      "eval_environment_accuracy": 0.980091723994163,
+      "eval_environment_f1": 0.9785258941308824,
+      "eval_instability_accuracy": 0.943089430894309,
+      "eval_instability_f1": 0.9426570245612531,
+      "eval_leadership_accuracy": 0.8357306649989577,
+      "eval_leadership_f1": 0.8319611907018966,
+      "eval_loss": 0.21395592391490936,
+      "eval_overall_accuracy": 0.9376433187408796,
+      "eval_overall_f1": 0.9341971854179941,
+      "eval_race_accuracy": 0.9546591619762351,
+      "eval_race_f1": 0.9516178104963666,
+      "eval_religion_accuracy": 0.9573691890765061,
+      "eval_religion_f1": 0.9546439142787715,
+      "eval_runtime": 10.9576,
+      "eval_safety_accuracy": 0.9206795914112987,
+      "eval_safety_f1": 0.9157232916504305,
+      "eval_samples_per_second": 875.555,
+      "eval_steps_per_second": 54.756,
+      "step": 7197
+    },
+    {
+      "epoch": 3.0012505210504377,
+      "grad_norm": 1.1601636409759521,
+      "learning_rate": 5.762564671101256e-05,
+      "loss": 0.1347,
+      "step": 7200
+    },
+    {
+      "epoch": 3.042934556065027,
+      "grad_norm": 2.3385512828826904,
+      "learning_rate": 5.744087213599408e-05,
+      "loss": 0.1002,
+      "step": 7300
+    },
+    {
+      "epoch": 3.0846185910796167,
+      "grad_norm": 0.9694398641586304,
+      "learning_rate": 5.7256097560975603e-05,
+      "loss": 0.0957,
+      "step": 7400
+    },
+    {
+      "epoch": 3.1263026260942057,
+      "grad_norm": 1.703196406364441,
+      "learning_rate": 5.707132298595713e-05,
+      "loss": 0.1002,
+      "step": 7500
+    },
+    {
+      "epoch": 3.1679866611087952,
+      "grad_norm": 2.1096134185791016,
+      "learning_rate": 5.6886548410938654e-05,
+      "loss": 0.1061,
+      "step": 7600
+    },
+    {
+      "epoch": 3.2096706961233847,
+      "grad_norm": 1.8544224500656128,
+      "learning_rate": 5.6701773835920166e-05,
+      "loss": 0.0964,
+      "step": 7700
+    },
+    {
+      "epoch": 3.2513547311379742,
+      "grad_norm": 1.9971802234649658,
+      "learning_rate": 5.651699926090169e-05,
+      "loss": 0.1036,
+      "step": 7800
+    },
+    {
+      "epoch": 3.2930387661525637,
+      "grad_norm": 1.9724899530410767,
+      "learning_rate": 5.6332224685883216e-05,
+      "loss": 0.0992,
+      "step": 7900
+    },
+    {
+      "epoch": 3.334722801167153,
+      "grad_norm": 3.2631657123565674,
+      "learning_rate": 5.614745011086474e-05,
+      "loss": 0.1022,
+      "step": 8000
+    },
+    {
+      "epoch": 3.3764068361817423,
+      "grad_norm": 1.6194812059402466,
+      "learning_rate": 5.596267553584627e-05,
+      "loss": 0.1005,
+      "step": 8100
+    },
+    {
+      "epoch": 3.418090871196332,
+      "grad_norm": 1.889183759689331,
+      "learning_rate": 5.577790096082778e-05,
+      "loss": 0.0981,
+      "step": 8200
+    },
+    {
+      "epoch": 3.4597749062109213,
+      "grad_norm": 1.4375004768371582,
+      "learning_rate": 5.5593126385809304e-05,
+      "loss": 0.1056,
+      "step": 8300
+    },
+    {
+      "epoch": 3.5014589412255104,
+      "grad_norm": 1.5392543077468872,
+      "learning_rate": 5.540835181079083e-05,
+      "loss": 0.1039,
+      "step": 8400
+    },
+    {
+      "epoch": 3.5431429762401,
+      "grad_norm": 2.7214481830596924,
+      "learning_rate": 5.5223577235772354e-05,
+      "loss": 0.0974,
+      "step": 8500
+    },
+    {
+      "epoch": 3.5848270112546894,
+      "grad_norm": 3.4187774658203125,
+      "learning_rate": 5.503880266075388e-05,
+      "loss": 0.1095,
+      "step": 8600
+    },
+    {
+      "epoch": 3.626511046269279,
+      "grad_norm": 2.25942063331604,
+      "learning_rate": 5.485402808573539e-05,
+      "loss": 0.0982,
+      "step": 8700
+    },
+    {
+      "epoch": 3.6681950812838684,
+      "grad_norm": 1.933835506439209,
+      "learning_rate": 5.4669253510716916e-05,
+      "loss": 0.1058,
+      "step": 8800
+    },
+    {
+      "epoch": 3.709879116298458,
+      "grad_norm": 1.7486668825149536,
+      "learning_rate": 5.448447893569844e-05,
+      "loss": 0.099,
+      "step": 8900
+    },
+    {
+      "epoch": 3.751563151313047,
+      "grad_norm": 1.5157849788665771,
+      "learning_rate": 5.429970436067997e-05,
+      "loss": 0.1014,
+      "step": 9000
+    },
+    {
+      "epoch": 3.7932471863276365,
+      "grad_norm": 2.798487663269043,
+      "learning_rate": 5.411492978566149e-05,
+      "loss": 0.1122,
+      "step": 9100
+    },
+    {
+      "epoch": 3.834931221342226,
+      "grad_norm": 1.6031073331832886,
+      "learning_rate": 5.393015521064302e-05,
+      "loss": 0.1142,
+      "step": 9200
+    },
+    {
+      "epoch": 3.8766152563568155,
+      "grad_norm": 1.2697032690048218,
+      "learning_rate": 5.374538063562453e-05,
+      "loss": 0.0987,
+      "step": 9300
+    },
+    {
+      "epoch": 3.9182992913714045,
+      "grad_norm": 2.0120253562927246,
+      "learning_rate": 5.3560606060606054e-05,
+      "loss": 0.0994,
+      "step": 9400
+    },
+    {
+      "epoch": 3.959983326385994,
+      "grad_norm": 1.9906800985336304,
+      "learning_rate": 5.337583148558758e-05,
+      "loss": 0.1022,
+      "step": 9500
+    },
+    {
+      "epoch": 4.0,
+      "eval_administration_accuracy": 0.9391286220554513,
+      "eval_administration_f1": 0.9314051851573208,
+      "eval_corruption_accuracy": 0.9487179487179487,
+      "eval_corruption_f1": 0.9492616863366112,
+      "eval_democracy_accuracy": 0.9540337711069419,
+      "eval_democracy_f1": 0.9486322890870622,
+      "eval_development_accuracy": 0.9184907233687721,
+      "eval_development_f1": 0.9117009476732075,
+      "eval_economy_accuracy": 0.9196372732958099,
+      "eval_economy_f1": 0.9214363800488883,
+      "eval_education_accuracy": 0.9672712111736502,
+      "eval_education_f1": 0.9642091841735867,
+      "eval_environment_accuracy": 0.9809255784865542,
+      "eval_environment_f1": 0.9795730735612355,
+      "eval_instability_accuracy": 0.9282885136543673,
+      "eval_instability_f1": 0.9323212368699592,
+      "eval_leadership_accuracy": 0.8347925786950178,
+      "eval_leadership_f1": 0.8300304516685039,
+      "eval_loss": 0.24011321365833282,
+      "eval_overall_accuracy": 0.9345250503787087,
+      "eval_overall_f1": 0.9324514217297923,
+      "eval_race_accuracy": 0.9515322076297686,
+      "eval_race_f1": 0.9511929125632914,
+      "eval_religion_accuracy": 0.9595580571190327,
+      "eval_religion_f1": 0.9582076782235817,
+      "eval_runtime": 10.9263,
+      "eval_safety_accuracy": 0.9119241192411924,
+      "eval_safety_f1": 0.9114460353942603,
+      "eval_samples_per_second": 878.068,
+      "eval_steps_per_second": 54.914,
+      "step": 9596
+    },
+    {
+      "epoch": 4.0016673614005835,
+      "grad_norm": 1.2328238487243652,
+      "learning_rate": 5.3191056910569105e-05,
+      "loss": 0.1003,
+      "step": 9600
+    },
+    {
+      "epoch": 4.043351396415173,
+      "grad_norm": 0.6457404494285583,
+      "learning_rate": 5.300628233555063e-05,
+      "loss": 0.0663,
+      "step": 9700
+    },
+    {
+      "epoch": 4.0850354314297626,
+      "grad_norm": 2.586268901824951,
+      "learning_rate": 5.282150776053214e-05,
+      "loss": 0.077,
+      "step": 9800
+    },
+    {
+      "epoch": 4.126719466444352,
+      "grad_norm": 2.192460060119629,
+      "learning_rate": 5.263673318551367e-05,
+      "loss": 0.0707,
+      "step": 9900
+    },
+    {
+      "epoch": 4.168403501458942,
+      "grad_norm": 0.8038670420646667,
+      "learning_rate": 5.245195861049519e-05,
+      "loss": 0.0697,
+      "step": 10000
+    },
+    {
+      "epoch": 4.210087536473531,
+      "grad_norm": 1.8463445901870728,
+      "learning_rate": 5.22690317812269e-05,
+      "loss": 0.0842,
+      "step": 10100
+    },
+    {
+      "epoch": 4.25177157148812,
+      "grad_norm": 1.395668387413025,
+      "learning_rate": 5.2084257206208424e-05,
+      "loss": 0.0862,
+      "step": 10200
+    },
+    {
+      "epoch": 4.293455606502709,
+      "grad_norm": 2.2199463844299316,
+      "learning_rate": 5.189948263118994e-05,
+      "loss": 0.0795,
+      "step": 10300
+    },
+    {
+      "epoch": 4.335139641517299,
+      "grad_norm": 1.7197673320770264,
+      "learning_rate": 5.171470805617147e-05,
+      "loss": 0.0784,
+      "step": 10400
+    },
+    {
+      "epoch": 4.376823676531888,
+      "grad_norm": 0.9125378131866455,
+      "learning_rate": 5.1529933481152986e-05,
+      "loss": 0.0875,
+      "step": 10500
+    },
+    {
+      "epoch": 4.418507711546478,
+      "grad_norm": 1.349757194519043,
+      "learning_rate": 5.134515890613451e-05,
+      "loss": 0.0778,
+      "step": 10600
+    },
+    {
+      "epoch": 4.460191746561067,
+      "grad_norm": 1.304445505142212,
+      "learning_rate": 5.1160384331116037e-05,
+      "loss": 0.0863,
+      "step": 10700
+    },
+    {
+      "epoch": 4.501875781575657,
+      "grad_norm": 1.316042184829712,
+      "learning_rate": 5.0975609756097555e-05,
+      "loss": 0.0758,
+      "step": 10800
+    },
+    {
+      "epoch": 4.543559816590246,
+      "grad_norm": 1.7575550079345703,
+      "learning_rate": 5.079083518107908e-05,
+      "loss": 0.079,
+      "step": 10900
+    },
+    {
+      "epoch": 4.585243851604836,
+      "grad_norm": 3.9689066410064697,
+      "learning_rate": 5.06060606060606e-05,
+      "loss": 0.0726,
+      "step": 11000
+    },
+    {
+      "epoch": 4.626927886619425,
+      "grad_norm": 3.6912765502929688,
+      "learning_rate": 5.0421286031042124e-05,
+      "loss": 0.0795,
+      "step": 11100
+    },
+    {
+      "epoch": 4.668611921634014,
+      "grad_norm": 2.2863271236419678,
+      "learning_rate": 5.023651145602365e-05,
+      "loss": 0.0796,
+      "step": 11200
+    },
+    {
+      "epoch": 4.710295956648603,
+      "grad_norm": 0.5653240084648132,
+      "learning_rate": 5.005173688100517e-05,
+      "loss": 0.0798,
+      "step": 11300
+    },
+    {
+      "epoch": 4.751979991663193,
+      "grad_norm": 0.5571924448013306,
+      "learning_rate": 4.986696230598669e-05,
+      "loss": 0.0784,
+      "step": 11400
+    },
+    {
+      "epoch": 4.793664026677782,
+      "grad_norm": 2.2439966201782227,
+      "learning_rate": 4.968218773096822e-05,
+      "loss": 0.0814,
+      "step": 11500
+    },
+    {
+      "epoch": 4.835348061692372,
+      "grad_norm": 3.0010180473327637,
+      "learning_rate": 4.949741315594974e-05,
+      "loss": 0.0744,
+      "step": 11600
+    },
+    {
+      "epoch": 4.877032096706961,
+      "grad_norm": 1.1905595064163208,
+      "learning_rate": 4.9312638580931255e-05,
+      "loss": 0.0801,
+      "step": 11700
+    },
+    {
+      "epoch": 4.918716131721551,
+      "grad_norm": 1.0284569263458252,
+      "learning_rate": 4.912786400591278e-05,
+      "loss": 0.0801,
+      "step": 11800
+    },
+    {
+      "epoch": 4.96040016673614,
+      "grad_norm": 1.0152745246887207,
+      "learning_rate": 4.8943089430894306e-05,
+      "loss": 0.0713,
+      "step": 11900
+    },
+    {
+      "epoch": 5.0,
+      "eval_administration_accuracy": 0.9377736085053158,
+      "eval_administration_f1": 0.9276824923430157,
+      "eval_corruption_accuracy": 0.9577861163227017,
+      "eval_corruption_f1": 0.9538795910231914,
+      "eval_democracy_accuracy": 0.9540337711069419,
+      "eval_democracy_f1": 0.9502590354742151,
+      "eval_development_accuracy": 0.9175526370648321,
+      "eval_development_f1": 0.9108745507581222,
+      "eval_economy_accuracy": 0.9300604544506984,
+      "eval_economy_f1": 0.9260225965314521,
+      "eval_education_accuracy": 0.9670627475505524,
+      "eval_education_f1": 0.9645423346801522,
+      "eval_environment_accuracy": 0.9807171148634564,
+      "eval_environment_f1": 0.978908255723529,
+      "eval_instability_accuracy": 0.9444444444444444,
+      "eval_instability_f1": 0.9428908884341405,
+      "eval_leadership_accuracy": 0.8395872420262664,
+      "eval_leadership_f1": 0.8349609938727663,
+      "eval_loss": 0.25315749645233154,
+      "eval_overall_accuracy": 0.9378865263011605,
+      "eval_overall_f1": 0.9340809877025569,
+      "eval_race_accuracy": 0.9501771940796331,
+      "eval_race_f1": 0.949342954475387,
+      "eval_religion_accuracy": 0.9579945799457995,
+      "eval_religion_f1": 0.9551951539975265,
+      "eval_runtime": 10.9532,
+      "eval_safety_accuracy": 0.9174484052532833,
+      "eval_safety_f1": 0.9144130051171864,
+      "eval_samples_per_second": 875.907,
+      "eval_steps_per_second": 54.778,
+      "step": 11995
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 38384,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 16,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 2,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 2
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.010440095766938e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 7e-05,
+    "num_train_epochs": 16
+  }
+}

run-3/checkpoint-11995/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2574f00335e03e693d14802bd78255c32eb346c265bc7cfbd1b32046ac22108
+size 5368

run-3/checkpoint-9596/config.json

@@ -0,0 +1,125 @@
+{
+  "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",
+    "5": "LABEL_5",
+    "6": "LABEL_6",
+    "7": "LABEL_7",
+    "8": "LABEL_8",
+    "9": "LABEL_9",
+    "10": "LABEL_10",
+    "11": "LABEL_11",
+    "12": "LABEL_12",
+    "13": "LABEL_13",
+    "14": "LABEL_14",
+    "15": "LABEL_15",
+    "16": "LABEL_16",
+    "17": "LABEL_17",
+    "18": "LABEL_18",
+    "19": "LABEL_19",
+    "20": "LABEL_20",
+    "21": "LABEL_21",
+    "22": "LABEL_22",
+    "23": "LABEL_23",
+    "24": "LABEL_24",
+    "25": "LABEL_25",
+    "26": "LABEL_26",
+    "27": "LABEL_27",
+    "28": "LABEL_28",
+    "29": "LABEL_29",
+    "30": "LABEL_30",
+    "31": "LABEL_31",
+    "32": "LABEL_32",
+    "33": "LABEL_33",
+    "34": "LABEL_34",
+    "35": "LABEL_35",
+    "36": "LABEL_36",
+    "37": "LABEL_37",
+    "38": "LABEL_38",
+    "39": "LABEL_39",
+    "40": "LABEL_40",
+    "41": "LABEL_41",
+    "42": "LABEL_42",
+    "43": "LABEL_43",
+    "44": "LABEL_44",
+    "45": "LABEL_45",
+    "46": "LABEL_46",
+    "47": "LABEL_47"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_10": 10,
+    "LABEL_11": 11,
+    "LABEL_12": 12,
+    "LABEL_13": 13,
+    "LABEL_14": 14,
+    "LABEL_15": 15,
+    "LABEL_16": 16,
+    "LABEL_17": 17,
+    "LABEL_18": 18,
+    "LABEL_19": 19,
+    "LABEL_2": 2,
+    "LABEL_20": 20,
+    "LABEL_21": 21,
+    "LABEL_22": 22,
+    "LABEL_23": 23,
+    "LABEL_24": 24,
+    "LABEL_25": 25,
+    "LABEL_26": 26,
+    "LABEL_27": 27,
+    "LABEL_28": 28,
+    "LABEL_29": 29,
+    "LABEL_3": 3,
+    "LABEL_30": 30,
+    "LABEL_31": 31,
+    "LABEL_32": 32,
+    "LABEL_33": 33,
+    "LABEL_34": 34,
+    "LABEL_35": 35,
+    "LABEL_36": 36,
+    "LABEL_37": 37,
+    "LABEL_38": 38,
+    "LABEL_39": 39,
+    "LABEL_4": 4,
+    "LABEL_40": 40,
+    "LABEL_41": 41,
+    "LABEL_42": 42,
+    "LABEL_43": 43,
+    "LABEL_44": 44,
+    "LABEL_45": 45,
+    "LABEL_46": 46,
+    "LABEL_47": 47,
+    "LABEL_5": 5,
+    "LABEL_6": 6,
+    "LABEL_7": 7,
+    "LABEL_8": 8,
+    "LABEL_9": 9
+  },
+  "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",
+  "torch_dtype": "float32",
+  "transformers_version": "4.50.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-3/checkpoint-9596/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:968a15cb31b41fe56989ccdd8440f972426e0a23047b33997c6aa4ea597d5cfd
+size 438100144

run-3/checkpoint-9596/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f3b9e52a380b2540fe23166319fa2d856b667e8675dd76a033030d7db3c6a9d
+size 876321402

run-3/checkpoint-9596/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ee8bdec2336829d8b665872a25fb80e042d5fa43fe24ad4b60b39cb823c692f
+size 14244

run-3/checkpoint-9596/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea98e376375c91bff3eebcb5729ce96ae5c9120c14d0fcbc2cbcefe83d613dfa
+size 988

run-3/checkpoint-9596/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc6b3745bb6c72eb1b207281776b9c9a1763bfefef4c532660eced81640f6ec0
+size 1064

run-3/checkpoint-9596/trainer_state.json

@@ -0,0 +1,848 @@
+{
+  "best_global_step": 7197,
+  "best_metric": 0.9341971854179941,
+  "best_model_checkpoint": "./results/run-3/checkpoint-7197",
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 9596,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.041684035014589414,
+      "grad_norm": 1.7545671463012695,
+      "learning_rate": 1.386e-05,
+      "loss": 1.2954,
+      "step": 100
+    },
+    {
+      "epoch": 0.08336807002917883,
+      "grad_norm": 0.7225435376167297,
+      "learning_rate": 2.7859999999999998e-05,
+      "loss": 0.5134,
+      "step": 200
+    },
+    {
+      "epoch": 0.12505210504376824,
+      "grad_norm": 0.9463444948196411,
+      "learning_rate": 4.1859999999999996e-05,
+      "loss": 0.4399,
+      "step": 300
+    },
+    {
+      "epoch": 0.16673614005835766,
+      "grad_norm": 0.8181287050247192,
+      "learning_rate": 5.586e-05,
+      "loss": 0.3608,
+      "step": 400
+    },
+    {
+      "epoch": 0.20842017507294705,
+      "grad_norm": 1.4074561595916748,
+      "learning_rate": 6.986e-05,
+      "loss": 0.3303,
+      "step": 500
+    },
+    {
+      "epoch": 0.25010421008753647,
+      "grad_norm": 1.232994556427002,
+      "learning_rate": 6.98170731707317e-05,
+      "loss": 0.3118,
+      "step": 600
+    },
+    {
+      "epoch": 0.29178824510212586,
+      "grad_norm": 1.1151797771453857,
+      "learning_rate": 6.96341463414634e-05,
+      "loss": 0.3067,
+      "step": 700
+    },
+    {
+      "epoch": 0.3334722801167153,
+      "grad_norm": 1.3453563451766968,
+      "learning_rate": 6.944937176644493e-05,
+      "loss": 0.2959,
+      "step": 800
+    },
+    {
+      "epoch": 0.3751563151313047,
+      "grad_norm": 1.3014110326766968,
+      "learning_rate": 6.926459719142646e-05,
+      "loss": 0.2803,
+      "step": 900
+    },
+    {
+      "epoch": 0.4168403501458941,
+      "grad_norm": 2.4783546924591064,
+      "learning_rate": 6.907982261640798e-05,
+      "loss": 0.275,
+      "step": 1000
+    },
+    {
+      "epoch": 0.45852438516048355,
+      "grad_norm": 1.0991694927215576,
+      "learning_rate": 6.88950480413895e-05,
+      "loss": 0.2725,
+      "step": 1100
+    },
+    {
+      "epoch": 0.5002084201750729,
+      "grad_norm": 1.8534690141677856,
+      "learning_rate": 6.871027346637102e-05,
+      "loss": 0.2965,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5418924551896623,
+      "grad_norm": 1.533751130104065,
+      "learning_rate": 6.852549889135254e-05,
+      "loss": 0.257,
+      "step": 1300
+    },
+    {
+      "epoch": 0.5835764902042517,
+      "grad_norm": 1.39555025100708,
+      "learning_rate": 6.834072431633407e-05,
+      "loss": 0.2666,
+      "step": 1400
+    },
+    {
+      "epoch": 0.6252605252188412,
+      "grad_norm": 1.1627668142318726,
+      "learning_rate": 6.81559497413156e-05,
+      "loss": 0.2461,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6669445602334306,
+      "grad_norm": 1.319166660308838,
+      "learning_rate": 6.79711751662971e-05,
+      "loss": 0.242,
+      "step": 1600
+    },
+    {
+      "epoch": 0.70862859524802,
+      "grad_norm": 1.3535823822021484,
+      "learning_rate": 6.778640059127863e-05,
+      "loss": 0.2463,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7503126302626094,
+      "grad_norm": 1.8609535694122314,
+      "learning_rate": 6.760162601626016e-05,
+      "loss": 0.2367,
+      "step": 1800
+    },
+    {
+      "epoch": 0.7919966652771988,
+      "grad_norm": 1.86128830909729,
+      "learning_rate": 6.741685144124168e-05,
+      "loss": 0.2449,
+      "step": 1900
+    },
+    {
+      "epoch": 0.8336807002917882,
+      "grad_norm": 1.8068938255310059,
+      "learning_rate": 6.723207686622321e-05,
+      "loss": 0.2329,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8753647353063777,
+      "grad_norm": 1.4341034889221191,
+      "learning_rate": 6.704730229120472e-05,
+      "loss": 0.2342,
+      "step": 2100
+    },
+    {
+      "epoch": 0.9170487703209671,
+      "grad_norm": 0.7307045459747314,
+      "learning_rate": 6.686252771618624e-05,
+      "loss": 0.2358,
+      "step": 2200
+    },
+    {
+      "epoch": 0.9587328053355565,
+      "grad_norm": 2.0308947563171387,
+      "learning_rate": 6.667775314116777e-05,
+      "loss": 0.2244,
+      "step": 2300
+    },
+    {
+      "epoch": 1.0,
+      "eval_administration_accuracy": 0.9317281634354805,
+      "eval_administration_f1": 0.9110301299328539,
+      "eval_corruption_accuracy": 0.9541380029184907,
+      "eval_corruption_f1": 0.9489574147600867,
+      "eval_democracy_accuracy": 0.9473629351678132,
+      "eval_democracy_f1": 0.935646230598331,
+      "eval_development_accuracy": 0.910464873879508,
+      "eval_development_f1": 0.9008646598644421,
+      "eval_economy_accuracy": 0.9235980821346675,
+      "eval_economy_f1": 0.9156600494130359,
+      "eval_education_accuracy": 0.9622680842193038,
+      "eval_education_f1": 0.9586233907920864,
+      "eval_environment_accuracy": 0.978319783197832,
+      "eval_environment_f1": 0.9770754665605412,
+      "eval_instability_accuracy": 0.9398582447362935,
+      "eval_instability_f1": 0.9299109706262821,
+      "eval_leadership_accuracy": 0.8081092349385033,
+      "eval_leadership_f1": 0.7818188134942813,
+      "eval_loss": 0.2264394313097,
+      "eval_overall_accuracy": 0.9310245987075256,
+      "eval_overall_f1": 0.9219705720705208,
+      "eval_race_accuracy": 0.9487179487179487,
+      "eval_race_f1": 0.9448562040025291,
+      "eval_religion_accuracy": 0.9527829893683553,
+      "eval_religion_f1": 0.950851997767183,
+      "eval_runtime": 11.1066,
+      "eval_safety_accuracy": 0.9149468417761101,
+      "eval_safety_f1": 0.9083515370345979,
+      "eval_samples_per_second": 863.809,
+      "eval_steps_per_second": 54.022,
+      "step": 2399
+    },
+    {
+      "epoch": 1.0004168403501459,
+      "grad_norm": 1.3034080266952515,
+      "learning_rate": 6.64929785661493e-05,
+      "loss": 0.2417,
+      "step": 2400
+    },
+    {
+      "epoch": 1.0421008753647354,
+      "grad_norm": 1.3565603494644165,
+      "learning_rate": 6.630820399113082e-05,
+      "loss": 0.1946,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0837849103793247,
+      "grad_norm": 2.098259449005127,
+      "learning_rate": 6.612342941611233e-05,
+      "loss": 0.1911,
+      "step": 2600
+    },
+    {
+      "epoch": 1.1254689453939142,
+      "grad_norm": 1.0764553546905518,
+      "learning_rate": 6.593865484109386e-05,
+      "loss": 0.1821,
+      "step": 2700
+    },
+    {
+      "epoch": 1.1671529804085035,
+      "grad_norm": 1.4594541788101196,
+      "learning_rate": 6.575388026607538e-05,
+      "loss": 0.1851,
+      "step": 2800
+    },
+    {
+      "epoch": 1.208837015423093,
+      "grad_norm": 1.8065540790557861,
+      "learning_rate": 6.556910569105691e-05,
+      "loss": 0.1929,
+      "step": 2900
+    },
+    {
+      "epoch": 1.2505210504376825,
+      "grad_norm": 0.8411226272583008,
+      "learning_rate": 6.538433111603843e-05,
+      "loss": 0.1882,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2922050854522718,
+      "grad_norm": 1.886057734489441,
+      "learning_rate": 6.519955654101994e-05,
+      "loss": 0.1861,
+      "step": 3100
+    },
+    {
+      "epoch": 1.3338891204668613,
+      "grad_norm": 1.2633222341537476,
+      "learning_rate": 6.501478196600147e-05,
+      "loss": 0.1816,
+      "step": 3200
+    },
+    {
+      "epoch": 1.3755731554814505,
+      "grad_norm": 2.1033339500427246,
+      "learning_rate": 6.4830007390983e-05,
+      "loss": 0.1887,
+      "step": 3300
+    },
+    {
+      "epoch": 1.41725719049604,
+      "grad_norm": 1.5505166053771973,
+      "learning_rate": 6.464523281596452e-05,
+      "loss": 0.1946,
+      "step": 3400
+    },
+    {
+      "epoch": 1.4589412255106295,
+      "grad_norm": 1.7376617193222046,
+      "learning_rate": 6.446045824094603e-05,
+      "loss": 0.1881,
+      "step": 3500
+    },
+    {
+      "epoch": 1.5006252605252188,
+      "grad_norm": 1.8713116645812988,
+      "learning_rate": 6.427568366592756e-05,
+      "loss": 0.1742,
+      "step": 3600
+    },
+    {
+      "epoch": 1.5423092955398081,
+      "grad_norm": 1.2346725463867188,
+      "learning_rate": 6.409090909090908e-05,
+      "loss": 0.1733,
+      "step": 3700
+    },
+    {
+      "epoch": 1.5839933305543976,
+      "grad_norm": 1.5446250438690186,
+      "learning_rate": 6.390613451589061e-05,
+      "loss": 0.1885,
+      "step": 3800
+    },
+    {
+      "epoch": 1.6256773655689871,
+      "grad_norm": 2.538613796234131,
+      "learning_rate": 6.372135994087213e-05,
+      "loss": 0.1737,
+      "step": 3900
+    },
+    {
+      "epoch": 1.6673614005835766,
+      "grad_norm": 1.1538047790527344,
+      "learning_rate": 6.353658536585365e-05,
+      "loss": 0.1836,
+      "step": 4000
+    },
+    {
+      "epoch": 1.709045435598166,
+      "grad_norm": 2.285994291305542,
+      "learning_rate": 6.335181079083517e-05,
+      "loss": 0.1934,
+      "step": 4100
+    },
+    {
+      "epoch": 1.7507294706127552,
+      "grad_norm": 1.3632721900939941,
+      "learning_rate": 6.31670362158167e-05,
+      "loss": 0.1792,
+      "step": 4200
+    },
+    {
+      "epoch": 1.7924135056273447,
+      "grad_norm": 1.3432073593139648,
+      "learning_rate": 6.298226164079822e-05,
+      "loss": 0.1665,
+      "step": 4300
+    },
+    {
+      "epoch": 1.8340975406419342,
+      "grad_norm": 4.086724758148193,
+      "learning_rate": 6.279748706577975e-05,
+      "loss": 0.1761,
+      "step": 4400
+    },
+    {
+      "epoch": 1.8757815756565237,
+      "grad_norm": 1.1902227401733398,
+      "learning_rate": 6.261271249076126e-05,
+      "loss": 0.1738,
+      "step": 4500
+    },
+    {
+      "epoch": 1.917465610671113,
+      "grad_norm": 2.1310391426086426,
+      "learning_rate": 6.242793791574278e-05,
+      "loss": 0.1822,
+      "step": 4600
+    },
+    {
+      "epoch": 1.9591496456857023,
+      "grad_norm": 1.384017825126648,
+      "learning_rate": 6.224316334072431e-05,
+      "loss": 0.1827,
+      "step": 4700
+    },
+    {
+      "epoch": 2.0,
+      "eval_administration_accuracy": 0.9382947675630603,
+      "eval_administration_f1": 0.93083241604641,
+      "eval_corruption_accuracy": 0.9589326662497394,
+      "eval_corruption_f1": 0.9564057119817679,
+      "eval_democracy_accuracy": 0.9522618303106108,
+      "eval_democracy_f1": 0.9459837872310805,
+      "eval_development_accuracy": 0.9119241192411924,
+      "eval_development_f1": 0.9077803592136469,
+      "eval_economy_accuracy": 0.9272461955388784,
+      "eval_economy_f1": 0.9256537091607892,
+      "eval_education_accuracy": 0.9613299979153638,
+      "eval_education_f1": 0.9608415429902122,
+      "eval_environment_accuracy": 0.978319783197832,
+      "eval_environment_f1": 0.9774790124713261,
+      "eval_instability_accuracy": 0.9247446320617052,
+      "eval_instability_f1": 0.9296973340071476,
+      "eval_leadership_accuracy": 0.8296852199291224,
+      "eval_leadership_f1": 0.8138352706726838,
+      "eval_loss": 0.21539363265037537,
+      "eval_overall_accuracy": 0.9336477659648392,
+      "eval_overall_f1": 0.9304371117477391,
+      "eval_race_accuracy": 0.9497602668334376,
+      "eval_race_f1": 0.9482550954258615,
+      "eval_religion_accuracy": 0.9559099437148217,
+      "eval_religion_f1": 0.9545949788562016,
+      "eval_runtime": 11.0609,
+      "eval_safety_accuracy": 0.9153637690223057,
+      "eval_safety_f1": 0.9138861229157411,
+      "eval_samples_per_second": 867.383,
+      "eval_steps_per_second": 54.245,
+      "step": 4798
+    },
+    {
+      "epoch": 2.0008336807002918,
+      "grad_norm": 0.8514364957809448,
+      "learning_rate": 6.205838876570583e-05,
+      "loss": 0.1822,
+      "step": 4800
+    },
+    {
+      "epoch": 2.0425177157148813,
+      "grad_norm": 1.2515586614608765,
+      "learning_rate": 6.187361419068736e-05,
+      "loss": 0.1312,
+      "step": 4900
+    },
+    {
+      "epoch": 2.084201750729471,
+      "grad_norm": 1.2924548387527466,
+      "learning_rate": 6.168883961566887e-05,
+      "loss": 0.1307,
+      "step": 5000
+    },
+    {
+      "epoch": 2.12588578574406,
+      "grad_norm": 1.7066774368286133,
+      "learning_rate": 6.15040650406504e-05,
+      "loss": 0.1248,
+      "step": 5100
+    },
+    {
+      "epoch": 2.1675698207586493,
+      "grad_norm": 1.4067167043685913,
+      "learning_rate": 6.131929046563192e-05,
+      "loss": 0.1318,
+      "step": 5200
+    },
+    {
+      "epoch": 2.209253855773239,
+      "grad_norm": 2.4648754596710205,
+      "learning_rate": 6.113451589061345e-05,
+      "loss": 0.1293,
+      "step": 5300
+    },
+    {
+      "epoch": 2.2509378907878284,
+      "grad_norm": 2.232621669769287,
+      "learning_rate": 6.0949741315594965e-05,
+      "loss": 0.128,
+      "step": 5400
+    },
+    {
+      "epoch": 2.292621925802418,
+      "grad_norm": 2.5116114616394043,
+      "learning_rate": 6.076496674057649e-05,
+      "loss": 0.1328,
+      "step": 5500
+    },
+    {
+      "epoch": 2.334305960817007,
+      "grad_norm": 1.2505563497543335,
+      "learning_rate": 6.0580192165558015e-05,
+      "loss": 0.1284,
+      "step": 5600
+    },
+    {
+      "epoch": 2.3759899958315964,
+      "grad_norm": 1.3521802425384521,
+      "learning_rate": 6.039541759053954e-05,
+      "loss": 0.1345,
+      "step": 5700
+    },
+    {
+      "epoch": 2.417674030846186,
+      "grad_norm": 1.0144108533859253,
+      "learning_rate": 6.021064301552106e-05,
+      "loss": 0.1366,
+      "step": 5800
+    },
+    {
+      "epoch": 2.4593580658607754,
+      "grad_norm": 0.8403565287590027,
+      "learning_rate": 6.002586844050258e-05,
+      "loss": 0.1397,
+      "step": 5900
+    },
+    {
+      "epoch": 2.501042100875365,
+      "grad_norm": 2.634355068206787,
+      "learning_rate": 5.98410938654841e-05,
+      "loss": 0.1373,
+      "step": 6000
+    },
+    {
+      "epoch": 2.542726135889954,
+      "grad_norm": 3.5822253227233887,
+      "learning_rate": 5.965631929046563e-05,
+      "loss": 0.1291,
+      "step": 6100
+    },
+    {
+      "epoch": 2.5844101709045435,
+      "grad_norm": 1.4991543292999268,
+      "learning_rate": 5.947154471544715e-05,
+      "loss": 0.1475,
+      "step": 6200
+    },
+    {
+      "epoch": 2.626094205919133,
+      "grad_norm": 1.9590423107147217,
+      "learning_rate": 5.928677014042867e-05,
+      "loss": 0.1411,
+      "step": 6300
+    },
+    {
+      "epoch": 2.6677782409337225,
+      "grad_norm": 1.2120991945266724,
+      "learning_rate": 5.910199556541019e-05,
+      "loss": 0.1397,
+      "step": 6400
+    },
+    {
+      "epoch": 2.7094622759483116,
+      "grad_norm": 2.270709991455078,
+      "learning_rate": 5.8917220990391715e-05,
+      "loss": 0.1398,
+      "step": 6500
+    },
+    {
+      "epoch": 2.751146310962901,
+      "grad_norm": 2.1711227893829346,
+      "learning_rate": 5.873244641537324e-05,
+      "loss": 0.1344,
+      "step": 6600
+    },
+    {
+      "epoch": 2.7928303459774906,
+      "grad_norm": 3.177982807159424,
+      "learning_rate": 5.854951958610495e-05,
+      "loss": 0.1377,
+      "step": 6700
+    },
+    {
+      "epoch": 2.83451438099208,
+      "grad_norm": 1.8457061052322388,
+      "learning_rate": 5.8364745011086465e-05,
+      "loss": 0.1277,
+      "step": 6800
+    },
+    {
+      "epoch": 2.8761984160066696,
+      "grad_norm": 1.9990291595458984,
+      "learning_rate": 5.817997043606799e-05,
+      "loss": 0.1311,
+      "step": 6900
+    },
+    {
+      "epoch": 2.917882451021259,
+      "grad_norm": 1.8258967399597168,
+      "learning_rate": 5.7995195861049516e-05,
+      "loss": 0.1359,
+      "step": 7000
+    },
+    {
+      "epoch": 2.959566486035848,
+      "grad_norm": 0.9961782097816467,
+      "learning_rate": 5.781042128603104e-05,
+      "loss": 0.1416,
+      "step": 7100
+    },
+    {
+      "epoch": 3.0,
+      "eval_administration_accuracy": 0.9374609130706691,
+      "eval_administration_f1": 0.9317222442324985,
+      "eval_corruption_accuracy": 0.9572649572649573,
+      "eval_corruption_f1": 0.9537446744289325,
+      "eval_democracy_accuracy": 0.9529914529914529,
+      "eval_democracy_f1": 0.9494686606905594,
+      "eval_development_accuracy": 0.9147383781530123,
+      "eval_development_f1": 0.9100235798889205,
+      "eval_economy_accuracy": 0.9329789451740671,
+      "eval_economy_f1": 0.9281892557992357,
+      "eval_education_accuracy": 0.964665415884928,
+      "eval_education_f1": 0.9620886841561839,
+      "eval_environment_accuracy": 0.980091723994163,
+      "eval_environment_f1": 0.9785258941308824,
+      "eval_instability_accuracy": 0.943089430894309,
+      "eval_instability_f1": 0.9426570245612531,
+      "eval_leadership_accuracy": 0.8357306649989577,
+      "eval_leadership_f1": 0.8319611907018966,
+      "eval_loss": 0.21395592391490936,
+      "eval_overall_accuracy": 0.9376433187408796,
+      "eval_overall_f1": 0.9341971854179941,
+      "eval_race_accuracy": 0.9546591619762351,
+      "eval_race_f1": 0.9516178104963666,
+      "eval_religion_accuracy": 0.9573691890765061,
+      "eval_religion_f1": 0.9546439142787715,
+      "eval_runtime": 10.9576,
+      "eval_safety_accuracy": 0.9206795914112987,
+      "eval_safety_f1": 0.9157232916504305,
+      "eval_samples_per_second": 875.555,
+      "eval_steps_per_second": 54.756,
+      "step": 7197
+    },
+    {
+      "epoch": 3.0012505210504377,
+      "grad_norm": 1.1601636409759521,
+      "learning_rate": 5.762564671101256e-05,
+      "loss": 0.1347,
+      "step": 7200
+    },
+    {
+      "epoch": 3.042934556065027,
+      "grad_norm": 2.3385512828826904,
+      "learning_rate": 5.744087213599408e-05,
+      "loss": 0.1002,
+      "step": 7300
+    },
+    {
+      "epoch": 3.0846185910796167,
+      "grad_norm": 0.9694398641586304,
+      "learning_rate": 5.7256097560975603e-05,
+      "loss": 0.0957,
+      "step": 7400
+    },
+    {
+      "epoch": 3.1263026260942057,
+      "grad_norm": 1.703196406364441,
+      "learning_rate": 5.707132298595713e-05,
+      "loss": 0.1002,
+      "step": 7500
+    },
+    {
+      "epoch": 3.1679866611087952,
+      "grad_norm": 2.1096134185791016,
+      "learning_rate": 5.6886548410938654e-05,
+      "loss": 0.1061,
+      "step": 7600
+    },
+    {
+      "epoch": 3.2096706961233847,
+      "grad_norm": 1.8544224500656128,
+      "learning_rate": 5.6701773835920166e-05,
+      "loss": 0.0964,
+      "step": 7700
+    },
+    {
+      "epoch": 3.2513547311379742,
+      "grad_norm": 1.9971802234649658,
+      "learning_rate": 5.651699926090169e-05,
+      "loss": 0.1036,
+      "step": 7800
+    },
+    {
+      "epoch": 3.2930387661525637,
+      "grad_norm": 1.9724899530410767,
+      "learning_rate": 5.6332224685883216e-05,
+      "loss": 0.0992,
+      "step": 7900
+    },
+    {
+      "epoch": 3.334722801167153,
+      "grad_norm": 3.2631657123565674,
+      "learning_rate": 5.614745011086474e-05,
+      "loss": 0.1022,
+      "step": 8000
+    },
+    {
+      "epoch": 3.3764068361817423,
+      "grad_norm": 1.6194812059402466,
+      "learning_rate": 5.596267553584627e-05,
+      "loss": 0.1005,
+      "step": 8100
+    },
+    {
+      "epoch": 3.418090871196332,
+      "grad_norm": 1.889183759689331,
+      "learning_rate": 5.577790096082778e-05,
+      "loss": 0.0981,
+      "step": 8200
+    },
+    {
+      "epoch": 3.4597749062109213,
+      "grad_norm": 1.4375004768371582,
+      "learning_rate": 5.5593126385809304e-05,
+      "loss": 0.1056,
+      "step": 8300
+    },
+    {
+      "epoch": 3.5014589412255104,
+      "grad_norm": 1.5392543077468872,
+      "learning_rate": 5.540835181079083e-05,
+      "loss": 0.1039,
+      "step": 8400
+    },
+    {
+      "epoch": 3.5431429762401,
+      "grad_norm": 2.7214481830596924,
+      "learning_rate": 5.5223577235772354e-05,
+      "loss": 0.0974,
+      "step": 8500
+    },
+    {
+      "epoch": 3.5848270112546894,
+      "grad_norm": 3.4187774658203125,
+      "learning_rate": 5.503880266075388e-05,
+      "loss": 0.1095,
+      "step": 8600
+    },
+    {
+      "epoch": 3.626511046269279,
+      "grad_norm": 2.25942063331604,
+      "learning_rate": 5.485402808573539e-05,
+      "loss": 0.0982,
+      "step": 8700
+    },
+    {
+      "epoch": 3.6681950812838684,
+      "grad_norm": 1.933835506439209,
+      "learning_rate": 5.4669253510716916e-05,
+      "loss": 0.1058,
+      "step": 8800
+    },
+    {
+      "epoch": 3.709879116298458,
+      "grad_norm": 1.7486668825149536,
+      "learning_rate": 5.448447893569844e-05,
+      "loss": 0.099,
+      "step": 8900
+    },
+    {
+      "epoch": 3.751563151313047,
+      "grad_norm": 1.5157849788665771,
+      "learning_rate": 5.429970436067997e-05,
+      "loss": 0.1014,
+      "step": 9000
+    },
+    {
+      "epoch": 3.7932471863276365,
+      "grad_norm": 2.798487663269043,
+      "learning_rate": 5.411492978566149e-05,
+      "loss": 0.1122,
+      "step": 9100
+    },
+    {
+      "epoch": 3.834931221342226,
+      "grad_norm": 1.6031073331832886,
+      "learning_rate": 5.393015521064302e-05,
+      "loss": 0.1142,
+      "step": 9200
+    },
+    {
+      "epoch": 3.8766152563568155,
+      "grad_norm": 1.2697032690048218,
+      "learning_rate": 5.374538063562453e-05,
+      "loss": 0.0987,
+      "step": 9300
+    },
+    {
+      "epoch": 3.9182992913714045,
+      "grad_norm": 2.0120253562927246,
+      "learning_rate": 5.3560606060606054e-05,
+      "loss": 0.0994,
+      "step": 9400
+    },
+    {
+      "epoch": 3.959983326385994,
+      "grad_norm": 1.9906800985336304,
+      "learning_rate": 5.337583148558758e-05,
+      "loss": 0.1022,
+      "step": 9500
+    },
+    {
+      "epoch": 4.0,
+      "eval_administration_accuracy": 0.9391286220554513,
+      "eval_administration_f1": 0.9314051851573208,
+      "eval_corruption_accuracy": 0.9487179487179487,
+      "eval_corruption_f1": 0.9492616863366112,
+      "eval_democracy_accuracy": 0.9540337711069419,
+      "eval_democracy_f1": 0.9486322890870622,
+      "eval_development_accuracy": 0.9184907233687721,
+      "eval_development_f1": 0.9117009476732075,
+      "eval_economy_accuracy": 0.9196372732958099,
+      "eval_economy_f1": 0.9214363800488883,
+      "eval_education_accuracy": 0.9672712111736502,
+      "eval_education_f1": 0.9642091841735867,
+      "eval_environment_accuracy": 0.9809255784865542,
+      "eval_environment_f1": 0.9795730735612355,
+      "eval_instability_accuracy": 0.9282885136543673,
+      "eval_instability_f1": 0.9323212368699592,
+      "eval_leadership_accuracy": 0.8347925786950178,
+      "eval_leadership_f1": 0.8300304516685039,
+      "eval_loss": 0.24011321365833282,
+      "eval_overall_accuracy": 0.9345250503787087,
+      "eval_overall_f1": 0.9324514217297923,
+      "eval_race_accuracy": 0.9515322076297686,
+      "eval_race_f1": 0.9511929125632914,
+      "eval_religion_accuracy": 0.9595580571190327,
+      "eval_religion_f1": 0.9582076782235817,
+      "eval_runtime": 10.9263,
+      "eval_safety_accuracy": 0.9119241192411924,
+      "eval_safety_f1": 0.9114460353942603,
+      "eval_samples_per_second": 878.068,
+      "eval_steps_per_second": 54.914,
+      "step": 9596
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 38384,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 16,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 2,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 1
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.999992218170163e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 7e-05,
+    "num_train_epochs": 16
+  }
+}

run-3/checkpoint-9596/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2574f00335e03e693d14802bd78255c32eb346c265bc7cfbd1b32046ac22108
+size 5368

run-4/checkpoint-600/config.json

@@ -0,0 +1,125 @@
+{
+  "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",
+    "5": "LABEL_5",
+    "6": "LABEL_6",
+    "7": "LABEL_7",
+    "8": "LABEL_8",
+    "9": "LABEL_9",
+    "10": "LABEL_10",
+    "11": "LABEL_11",
+    "12": "LABEL_12",
+    "13": "LABEL_13",
+    "14": "LABEL_14",
+    "15": "LABEL_15",
+    "16": "LABEL_16",
+    "17": "LABEL_17",
+    "18": "LABEL_18",
+    "19": "LABEL_19",
+    "20": "LABEL_20",
+    "21": "LABEL_21",
+    "22": "LABEL_22",
+    "23": "LABEL_23",
+    "24": "LABEL_24",
+    "25": "LABEL_25",
+    "26": "LABEL_26",
+    "27": "LABEL_27",
+    "28": "LABEL_28",
+    "29": "LABEL_29",
+    "30": "LABEL_30",
+    "31": "LABEL_31",
+    "32": "LABEL_32",
+    "33": "LABEL_33",
+    "34": "LABEL_34",
+    "35": "LABEL_35",
+    "36": "LABEL_36",
+    "37": "LABEL_37",
+    "38": "LABEL_38",
+    "39": "LABEL_39",
+    "40": "LABEL_40",
+    "41": "LABEL_41",
+    "42": "LABEL_42",
+    "43": "LABEL_43",
+    "44": "LABEL_44",
+    "45": "LABEL_45",
+    "46": "LABEL_46",
+    "47": "LABEL_47"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_10": 10,
+    "LABEL_11": 11,
+    "LABEL_12": 12,
+    "LABEL_13": 13,
+    "LABEL_14": 14,
+    "LABEL_15": 15,
+    "LABEL_16": 16,
+    "LABEL_17": 17,
+    "LABEL_18": 18,
+    "LABEL_19": 19,
+    "LABEL_2": 2,
+    "LABEL_20": 20,
+    "LABEL_21": 21,
+    "LABEL_22": 22,
+    "LABEL_23": 23,
+    "LABEL_24": 24,
+    "LABEL_25": 25,
+    "LABEL_26": 26,
+    "LABEL_27": 27,
+    "LABEL_28": 28,
+    "LABEL_29": 29,
+    "LABEL_3": 3,
+    "LABEL_30": 30,
+    "LABEL_31": 31,
+    "LABEL_32": 32,
+    "LABEL_33": 33,
+    "LABEL_34": 34,
+    "LABEL_35": 35,
+    "LABEL_36": 36,
+    "LABEL_37": 37,
+    "LABEL_38": 38,
+    "LABEL_39": 39,
+    "LABEL_4": 4,
+    "LABEL_40": 40,
+    "LABEL_41": 41,
+    "LABEL_42": 42,
+    "LABEL_43": 43,
+    "LABEL_44": 44,
+    "LABEL_45": 45,
+    "LABEL_46": 46,
+    "LABEL_47": 47,
+    "LABEL_5": 5,
+    "LABEL_6": 6,
+    "LABEL_7": 7,
+    "LABEL_8": 8,
+    "LABEL_9": 9
+  },
+  "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",
+  "torch_dtype": "float32",
+  "transformers_version": "4.50.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-4/checkpoint-600/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97687f5fed50ebc5579f2887c4a528f3caa14d8822166df6c145698780fc26c1
+size 438100144

run-4/checkpoint-600/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dada3e79362a6cf6e4c5af96391bc5057562c5fe878cd0d44eb2df7ab06cb2bf
+size 876321402

run-4/checkpoint-600/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03f6521411fa3fb31926db64cfd06b37e9d7c0607f2eec0bf790c73df4910854
+size 14244

run-4/checkpoint-600/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e32b8db4f4cba2b7b0157ab60a064591c3802a9067529d5fddbb1c6c1e99660b
+size 988

run-4/checkpoint-600/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe4dcc838fff3ee55425e473f2914dcb1cd2b4ae51a97d05940b655518d9727b
+size 1064

run-4/checkpoint-600/trainer_state.json

@@ -0,0 +1,123 @@
+{
+  "best_global_step": 600,
+  "best_metric": 0.8968188861118868,
+  "best_model_checkpoint": "./results/run-4/checkpoint-600",
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 600,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16673614005835766,
+      "grad_norm": 1.7758623361587524,
+      "learning_rate": 1e-05,
+      "loss": 1.3361,
+      "step": 100
+    },
+    {
+      "epoch": 0.3334722801167153,
+      "grad_norm": 0.6607375144958496,
+      "learning_rate": 2e-05,
+      "loss": 0.5099,
+      "step": 200
+    },
+    {
+      "epoch": 0.5002084201750729,
+      "grad_norm": 0.5690075159072876,
+      "learning_rate": 3e-05,
+      "loss": 0.4223,
+      "step": 300
+    },
+    {
+      "epoch": 0.6669445602334306,
+      "grad_norm": 0.848582923412323,
+      "learning_rate": 4e-05,
+      "loss": 0.3466,
+      "step": 400
+    },
+    {
+      "epoch": 0.8336807002917882,
+      "grad_norm": 0.7310414910316467,
+      "learning_rate": 5e-05,
+      "loss": 0.3032,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.5762994289398193,
+      "learning_rate": 4.7995991983967934e-05,
+      "loss": 0.2757,
+      "step": 600
+    },
+    {
+      "epoch": 1.0,
+      "eval_administration_accuracy": 0.9163018553262455,
+      "eval_administration_f1": 0.8799341161157713,
+      "eval_corruption_accuracy": 0.9427767354596623,
+      "eval_corruption_f1": 0.9297320702956003,
+      "eval_democracy_accuracy": 0.9418386491557224,
+      "eval_democracy_f1": 0.9228010908644549,
+      "eval_development_accuracy": 0.8867000208463623,
+      "eval_development_f1": 0.8490864044420574,
+      "eval_economy_accuracy": 0.9122368146758391,
+      "eval_economy_f1": 0.8967581427500766,
+      "eval_education_accuracy": 0.9583072753804461,
+      "eval_education_f1": 0.9515050567699045,
+      "eval_environment_accuracy": 0.9754012924744632,
+      "eval_environment_f1": 0.971384333060466,
+      "eval_instability_accuracy": 0.924119241192412,
+      "eval_instability_f1": 0.8935408759758452,
+      "eval_leadership_accuracy": 0.7805920366895976,
+      "eval_leadership_f1": 0.7320671383079426,
+      "eval_loss": 0.26127132773399353,
+      "eval_overall_accuracy": 0.9182127718713086,
+      "eval_overall_f1": 0.8968188861118868,
+      "eval_race_accuracy": 0.9360016677089847,
+      "eval_race_f1": 0.9209694401775296,
+      "eval_religion_accuracy": 0.9441317490097978,
+      "eval_religion_f1": 0.9328887566718251,
+      "eval_runtime": 10.8983,
+      "eval_safety_accuracy": 0.9001459245361685,
+      "eval_safety_f1": 0.8811592079111662,
+      "eval_samples_per_second": 880.318,
+      "eval_steps_per_second": 55.054,
+      "step": 600
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 2995,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 2,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.0100267260379136e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "gradient_accumulation_steps": 4,
+    "learning_rate": 5e-05,
+    "num_train_epochs": 5
+  }
+}

run-4/checkpoint-600/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d23c4b505069b659bf843cf83d2a0a253f99487d90fb9c81ceef3ac1152d69eb
+size 5368

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2574f00335e03e693d14802bd78255c32eb346c265bc7cfbd1b32046ac22108
+oid sha256:d23c4b505069b659bf843cf83d2a0a253f99487d90fb9c81ceef3ac1152d69eb
 size 5368