Training in progress, epoch 3

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f04ca5313e42aad7488aa63b678ef2efcfc6d1f482d81404f093322d0b4045d
+oid sha256:2aae3703a8dbd59e3bbe54c559f9ac85e6856a022e5aaa47f0bd29e64d1a6e97
 size 438100144

run-2/checkpoint-3600/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-2/checkpoint-3600/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2aae3703a8dbd59e3bbe54c559f9ac85e6856a022e5aaa47f0bd29e64d1a6e97
+size 438100144

run-2/checkpoint-3600/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a306fbfda27669d661212bcc235db77a2852d71e62029c9458875368fabca7
+size 876321402

run-2/checkpoint-3600/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:714a982d29c5cad10f1f61ff55536b945838337379aaedf05d2dafc18b602ef9
+size 14244

run-2/checkpoint-3600/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14aa284785350088ae14f96098e3480d199adb570a5ad5a358fcaedfe58e4e40
+size 988

run-2/checkpoint-3600/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01adb59376ee1f92e6540654bf2ee76b8ec1b5fea4c34c521a3ce1001ad5edef
+size 1064

run-2/checkpoint-3600/trainer_state.json

@@ -0,0 +1,401 @@
+{
+  "best_global_step": 2400,
+  "best_metric": 0.9339233416905627,
+  "best_model_checkpoint": "./results/run-2/checkpoint-2400",
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 3600,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.08336807002917883,
+      "grad_norm": 1.2878578901290894,
+      "learning_rate": 1.4e-05,
+      "loss": 1.2644,
+      "step": 100
+    },
+    {
+      "epoch": 0.16673614005835766,
+      "grad_norm": 1.0009047985076904,
+      "learning_rate": 2.8e-05,
+      "loss": 0.4758,
+      "step": 200
+    },
+    {
+      "epoch": 0.25010421008753647,
+      "grad_norm": 0.7635347843170166,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4054,
+      "step": 300
+    },
+    {
+      "epoch": 0.3334722801167153,
+      "grad_norm": 0.8168604373931885,
+      "learning_rate": 5.6e-05,
+      "loss": 0.3468,
+      "step": 400
+    },
+    {
+      "epoch": 0.4168403501458941,
+      "grad_norm": 1.1069319248199463,
+      "learning_rate": 7e-05,
+      "loss": 0.3051,
+      "step": 500
+    },
+    {
+      "epoch": 0.5002084201750729,
+      "grad_norm": 1.015086054801941,
+      "learning_rate": 6.962534789124383e-05,
+      "loss": 0.3044,
+      "step": 600
+    },
+    {
+      "epoch": 0.5835764902042517,
+      "grad_norm": 0.8430781364440918,
+      "learning_rate": 6.925069578248769e-05,
+      "loss": 0.2742,
+      "step": 700
+    },
+    {
+      "epoch": 0.6669445602334306,
+      "grad_norm": 0.950409471988678,
+      "learning_rate": 6.887604367373152e-05,
+      "loss": 0.2573,
+      "step": 800
+    },
+    {
+      "epoch": 0.7503126302626094,
+      "grad_norm": 1.171885371208191,
+      "learning_rate": 6.850139156497538e-05,
+      "loss": 0.2517,
+      "step": 900
+    },
+    {
+      "epoch": 0.8336807002917882,
+      "grad_norm": 0.9133914709091187,
+      "learning_rate": 6.812673945621922e-05,
+      "loss": 0.2435,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9170487703209671,
+      "grad_norm": 0.7939682602882385,
+      "learning_rate": 6.775208734746307e-05,
+      "loss": 0.2373,
+      "step": 1100
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.49670639634132385,
+      "learning_rate": 6.737743523870691e-05,
+      "loss": 0.2363,
+      "step": 1200
+    },
+    {
+      "epoch": 1.0,
+      "eval_administration_accuracy": 0.9302689180737961,
+      "eval_administration_f1": 0.9077189587684091,
+      "eval_corruption_accuracy": 0.9534083802376485,
+      "eval_corruption_f1": 0.9474431988592255,
+      "eval_democracy_accuracy": 0.9473629351678132,
+      "eval_democracy_f1": 0.9358042521135516,
+      "eval_development_accuracy": 0.9094225557640192,
+      "eval_development_f1": 0.8937523747915405,
+      "eval_economy_accuracy": 0.924119241192412,
+      "eval_economy_f1": 0.9156575066932242,
+      "eval_education_accuracy": 0.9621638524077548,
+      "eval_education_f1": 0.9583809787976314,
+      "eval_environment_accuracy": 0.9775901605169898,
+      "eval_environment_f1": 0.9754739686963207,
+      "eval_instability_accuracy": 0.9422555764019178,
+      "eval_instability_f1": 0.9343996801103537,
+      "eval_leadership_accuracy": 0.7970606629143214,
+      "eval_leadership_f1": 0.7702375495381067,
+      "eval_loss": 0.22745274007320404,
+      "eval_overall_accuracy": 0.9296261552359112,
+      "eval_overall_f1": 0.9199542168750479,
+      "eval_race_accuracy": 0.9505941213258287,
+      "eval_race_f1": 0.9483610849847235,
+      "eval_religion_accuracy": 0.9505941213258287,
+      "eval_religion_f1": 0.9489643730683611,
+      "eval_runtime": 11.0331,
+      "eval_safety_accuracy": 0.9106733375026058,
+      "eval_safety_f1": 0.9032566760791263,
+      "eval_samples_per_second": 869.566,
+      "eval_steps_per_second": 54.382,
+      "step": 1200
+    },
+    {
+      "epoch": 1.0833680700291788,
+      "grad_norm": 0.949184238910675,
+      "learning_rate": 6.700278312995076e-05,
+      "loss": 0.1968,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1667361400583576,
+      "grad_norm": 0.7735128402709961,
+      "learning_rate": 6.66281310211946e-05,
+      "loss": 0.1924,
+      "step": 1400
+    },
+    {
+      "epoch": 1.2501042100875366,
+      "grad_norm": 1.0187016725540161,
+      "learning_rate": 6.625347891243845e-05,
+      "loss": 0.1945,
+      "step": 1500
+    },
+    {
+      "epoch": 1.3334722801167154,
+      "grad_norm": 0.9707878828048706,
+      "learning_rate": 6.587882680368229e-05,
+      "loss": 0.1931,
+      "step": 1600
+    },
+    {
+      "epoch": 1.4168403501458942,
+      "grad_norm": 1.4602934122085571,
+      "learning_rate": 6.550417469492613e-05,
+      "loss": 0.1967,
+      "step": 1700
+    },
+    {
+      "epoch": 1.500208420175073,
+      "grad_norm": 1.2449183464050293,
+      "learning_rate": 6.512952258616998e-05,
+      "loss": 0.1843,
+      "step": 1800
+    },
+    {
+      "epoch": 1.5835764902042517,
+      "grad_norm": 1.4943019151687622,
+      "learning_rate": 6.475487047741382e-05,
+      "loss": 0.1844,
+      "step": 1900
+    },
+    {
+      "epoch": 1.6669445602334307,
+      "grad_norm": 1.8496220111846924,
+      "learning_rate": 6.438021836865767e-05,
+      "loss": 0.1797,
+      "step": 2000
+    },
+    {
+      "epoch": 1.7503126302626093,
+      "grad_norm": 1.0578210353851318,
+      "learning_rate": 6.400556625990151e-05,
+      "loss": 0.1875,
+      "step": 2100
+    },
+    {
+      "epoch": 1.8336807002917883,
+      "grad_norm": 1.0538480281829834,
+      "learning_rate": 6.363091415114536e-05,
+      "loss": 0.1685,
+      "step": 2200
+    },
+    {
+      "epoch": 1.917048770320967,
+      "grad_norm": 1.5351332426071167,
+      "learning_rate": 6.32562620423892e-05,
+      "loss": 0.1744,
+      "step": 2300
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.9578828811645508,
+      "learning_rate": 6.288160993363306e-05,
+      "loss": 0.1811,
+      "step": 2400
+    },
+    {
+      "epoch": 2.0,
+      "eval_administration_accuracy": 0.9344381905357515,
+      "eval_administration_f1": 0.9296143144580357,
+      "eval_corruption_accuracy": 0.9604961434229727,
+      "eval_corruption_f1": 0.9577803781696625,
+      "eval_democracy_accuracy": 0.9534083802376485,
+      "eval_democracy_f1": 0.9434755362160818,
+      "eval_development_accuracy": 0.9193245778611632,
+      "eval_development_f1": 0.9132597892423039,
+      "eval_economy_accuracy": 0.931311236189285,
+      "eval_economy_f1": 0.92884213698996,
+      "eval_education_accuracy": 0.9653950385657702,
+      "eval_education_f1": 0.963862144635937,
+      "eval_environment_accuracy": 0.979153637690223,
+      "eval_environment_f1": 0.9783610154594226,
+      "eval_instability_accuracy": 0.9459036898061288,
+      "eval_instability_f1": 0.9423381513536221,
+      "eval_leadership_accuracy": 0.8387533875338753,
+      "eval_leadership_f1": 0.8323519207670947,
+      "eval_loss": 0.2031705528497696,
+      "eval_overall_accuracy": 0.9372177055103883,
+      "eval_overall_f1": 0.9339233416905627,
+      "eval_race_accuracy": 0.9516364394413175,
+      "eval_race_f1": 0.9498984068182696,
+      "eval_religion_accuracy": 0.9558057119032729,
+      "eval_religion_f1": 0.9554428854878803,
+      "eval_runtime": 11.0378,
+      "eval_safety_accuracy": 0.9109860329372524,
+      "eval_safety_f1": 0.9118534206884823,
+      "eval_samples_per_second": 869.197,
+      "eval_steps_per_second": 54.359,
+      "step": 2400
+    },
+    {
+      "epoch": 2.083368070029179,
+      "grad_norm": 1.0708891153335571,
+      "learning_rate": 6.25069578248769e-05,
+      "loss": 0.1338,
+      "step": 2500
+    },
+    {
+      "epoch": 2.1667361400583576,
+      "grad_norm": 1.0800749063491821,
+      "learning_rate": 6.213230571612073e-05,
+      "loss": 0.1268,
+      "step": 2600
+    },
+    {
+      "epoch": 2.2501042100875366,
+      "grad_norm": 1.3577998876571655,
+      "learning_rate": 6.175765360736459e-05,
+      "loss": 0.1226,
+      "step": 2700
+    },
+    {
+      "epoch": 2.333472280116715,
+      "grad_norm": 1.325534701347351,
+      "learning_rate": 6.138300149860843e-05,
+      "loss": 0.1258,
+      "step": 2800
+    },
+    {
+      "epoch": 2.416840350145894,
+      "grad_norm": 1.7846063375473022,
+      "learning_rate": 6.100834938985228e-05,
+      "loss": 0.1308,
+      "step": 2900
+    },
+    {
+      "epoch": 2.500208420175073,
+      "grad_norm": 1.61028254032135,
+      "learning_rate": 6.0633697281096124e-05,
+      "loss": 0.1305,
+      "step": 3000
+    },
+    {
+      "epoch": 2.5835764902042517,
+      "grad_norm": 1.2228693962097168,
+      "learning_rate": 6.025904517233996e-05,
+      "loss": 0.1315,
+      "step": 3100
+    },
+    {
+      "epoch": 2.6669445602334307,
+      "grad_norm": 0.9486532807350159,
+      "learning_rate": 5.988439306358381e-05,
+      "loss": 0.1337,
+      "step": 3200
+    },
+    {
+      "epoch": 2.7503126302626093,
+      "grad_norm": 1.3199305534362793,
+      "learning_rate": 5.9509740954827654e-05,
+      "loss": 0.132,
+      "step": 3300
+    },
+    {
+      "epoch": 2.8336807002917883,
+      "grad_norm": 1.8821613788604736,
+      "learning_rate": 5.91350888460715e-05,
+      "loss": 0.1258,
+      "step": 3400
+    },
+    {
+      "epoch": 2.9170487703209673,
+      "grad_norm": 1.2816129922866821,
+      "learning_rate": 5.8760436737315346e-05,
+      "loss": 0.1282,
+      "step": 3500
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 5.737521171569824,
+      "learning_rate": 5.838578462855919e-05,
+      "loss": 0.1297,
+      "step": 3600
+    },
+    {
+      "epoch": 3.0,
+      "eval_administration_accuracy": 0.9375651448822181,
+      "eval_administration_f1": 0.9309689641816119,
+      "eval_corruption_accuracy": 0.9451740671252866,
+      "eval_corruption_f1": 0.9476001634812502,
+      "eval_democracy_accuracy": 0.9508025849489264,
+      "eval_democracy_f1": 0.9508783301626133,
+      "eval_development_accuracy": 0.921304982280592,
+      "eval_development_f1": 0.9145210502797096,
+      "eval_economy_accuracy": 0.930373149885345,
+      "eval_economy_f1": 0.9257602899924291,
+      "eval_education_accuracy": 0.9670627475505524,
+      "eval_education_f1": 0.9651765113114141,
+      "eval_environment_accuracy": 0.9805086512403586,
+      "eval_environment_f1": 0.9785021531425564,
+      "eval_instability_accuracy": 0.9419428809672712,
+      "eval_instability_f1": 0.9412425903038227,
+      "eval_leadership_accuracy": 0.8284344381905357,
+      "eval_leadership_f1": 0.8287745355631846,
+      "eval_loss": 0.21378232538700104,
+      "eval_overall_accuracy": 0.9360450976304634,
+      "eval_overall_f1": 0.9337781089602167,
+      "eval_race_accuracy": 0.9541380029184907,
+      "eval_race_f1": 0.9510004306979282,
+      "eval_religion_accuracy": 0.955701480091724,
+      "eval_religion_f1": 0.9539633431565642,
+      "eval_runtime": 11.1303,
+      "eval_safety_accuracy": 0.919533041484261,
+      "eval_safety_f1": 0.9169489452495161,
+      "eval_samples_per_second": 861.972,
+      "eval_steps_per_second": 53.907,
+      "step": 3600
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 19184,
+  "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": 3.030080178113741e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "gradient_accumulation_steps": 2,
+    "learning_rate": 7e-05,
+    "num_train_epochs": 16
+  }
+}

run-2/checkpoint-3600/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a417688ba3a38960888d219af94bafd1d9551df9005996513f2ac72dfc76c145
+size 5368