trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 50,
  "global_step": 1715,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "eval_accuracy": 0.35737704918032787,
      "eval_f1": 0.07522429261559696,
      "eval_loss": 1.667272925376892,
      "eval_precision": 0.05105386416861827,
      "eval_recall": 0.14285714285714285,
      "eval_runtime": 25.9688,
      "eval_samples_per_second": 11.745,
      "eval_steps_per_second": 1.502,
      "step": 50
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.5016393442622951,
      "eval_f1": 0.2412952804098472,
      "eval_loss": 1.472964882850647,
      "eval_precision": 0.22701762523191094,
      "eval_recall": 0.2760011649919907,
      "eval_runtime": 25.1854,
      "eval_samples_per_second": 12.11,
      "eval_steps_per_second": 1.549,
      "step": 100
    },
    {
      "epoch": 0.44,
      "eval_accuracy": 0.5672131147540984,
      "eval_f1": 0.35741119048011755,
      "eval_loss": 1.2623624801635742,
      "eval_precision": 0.46281513214918163,
      "eval_recall": 0.34605526990848096,
      "eval_runtime": 24.7658,
      "eval_samples_per_second": 12.315,
      "eval_steps_per_second": 1.575,
      "step": 150
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.5639344262295082,
      "eval_f1": 0.3512701619659464,
      "eval_loss": 1.2795928716659546,
      "eval_precision": 0.48283711313562055,
      "eval_recall": 0.34593069478390576,
      "eval_runtime": 26.4061,
      "eval_samples_per_second": 11.55,
      "eval_steps_per_second": 1.477,
      "step": 200
    },
    {
      "epoch": 0.73,
      "eval_accuracy": 0.639344262295082,
      "eval_f1": 0.42203695116777995,
      "eval_loss": 1.0558925867080688,
      "eval_precision": 0.48377222428420436,
      "eval_recall": 0.4281083079248217,
      "eval_runtime": 24.1696,
      "eval_samples_per_second": 12.619,
      "eval_steps_per_second": 1.614,
      "step": 250
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.659016393442623,
      "eval_f1": 0.5624420508569995,
      "eval_loss": 0.9904302954673767,
      "eval_precision": 0.6148969264343321,
      "eval_recall": 0.5786355212960718,
      "eval_runtime": 24.8748,
      "eval_samples_per_second": 12.261,
      "eval_steps_per_second": 1.568,
      "step": 300
    },
    {
      "epoch": 1.02,
      "eval_accuracy": 0.5967213114754099,
      "eval_f1": 0.42053130606223654,
      "eval_loss": 1.1112089157104492,
      "eval_precision": 0.5042065435600455,
      "eval_recall": 0.42027804344318104,
      "eval_runtime": 26.735,
      "eval_samples_per_second": 11.408,
      "eval_steps_per_second": 1.459,
      "step": 350
    },
    {
      "epoch": 1.17,
      "eval_accuracy": 0.6622950819672131,
      "eval_f1": 0.540005974333324,
      "eval_loss": 1.0169768333435059,
      "eval_precision": 0.6707201514365236,
      "eval_recall": 0.5058656801776068,
      "eval_runtime": 24.681,
      "eval_samples_per_second": 12.358,
      "eval_steps_per_second": 1.58,
      "step": 400
    },
    {
      "epoch": 1.31,
      "eval_accuracy": 0.7049180327868853,
      "eval_f1": 0.5548474859699348,
      "eval_loss": 0.963288426399231,
      "eval_precision": 0.6737802840434419,
      "eval_recall": 0.5320264412007532,
      "eval_runtime": 85.8855,
      "eval_samples_per_second": 3.551,
      "eval_steps_per_second": 0.454,
      "step": 450
    },
    {
      "epoch": 1.46,
      "grad_norm": 6.387063980102539,
      "learning_rate": 3.542274052478135e-05,
      "loss": 1.2491,
      "step": 500
    },
    {
      "epoch": 1.46,
      "eval_accuracy": 0.740983606557377,
      "eval_f1": 0.6032578704144542,
      "eval_loss": 0.8153039216995239,
      "eval_precision": 0.6389329389702317,
      "eval_recall": 0.5814238802770912,
      "eval_runtime": 25.7584,
      "eval_samples_per_second": 11.841,
      "eval_steps_per_second": 1.514,
      "step": 500
    },
    {
      "epoch": 1.6,
      "eval_accuracy": 0.7311475409836066,
      "eval_f1": 0.59968879894362,
      "eval_loss": 0.8648873567581177,
      "eval_precision": 0.7235630919034282,
      "eval_recall": 0.5553548759970778,
      "eval_runtime": 26.1622,
      "eval_samples_per_second": 11.658,
      "eval_steps_per_second": 1.491,
      "step": 550
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.780327868852459,
      "eval_f1": 0.6441369205133565,
      "eval_loss": 0.7085286974906921,
      "eval_precision": 0.6888112350883506,
      "eval_recall": 0.6200901173378238,
      "eval_runtime": 25.1463,
      "eval_samples_per_second": 12.129,
      "eval_steps_per_second": 1.551,
      "step": 600
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.7639344262295082,
      "eval_f1": 0.6348228528062524,
      "eval_loss": 0.778355062007904,
      "eval_precision": 0.6448629451780713,
      "eval_recall": 0.6364373118960275,
      "eval_runtime": 25.0645,
      "eval_samples_per_second": 12.169,
      "eval_steps_per_second": 1.556,
      "step": 650
    },
    {
      "epoch": 2.04,
      "eval_accuracy": 0.7475409836065574,
      "eval_f1": 0.6299370987134999,
      "eval_loss": 0.7480295896530151,
      "eval_precision": 0.6783989867133294,
      "eval_recall": 0.6207732331585543,
      "eval_runtime": 24.3631,
      "eval_samples_per_second": 12.519,
      "eval_steps_per_second": 1.601,
      "step": 700
    },
    {
      "epoch": 2.19,
      "eval_accuracy": 0.7672131147540984,
      "eval_f1": 0.6541376568128562,
      "eval_loss": 0.7434977293014526,
      "eval_precision": 0.7108263630655515,
      "eval_recall": 0.6220120665074793,
      "eval_runtime": 25.2918,
      "eval_samples_per_second": 12.059,
      "eval_steps_per_second": 1.542,
      "step": 750
    },
    {
      "epoch": 2.33,
      "eval_accuracy": 0.7311475409836066,
      "eval_f1": 0.6427856362037365,
      "eval_loss": 0.9037203192710876,
      "eval_precision": 0.7008431166108425,
      "eval_recall": 0.6458018130036478,
      "eval_runtime": 24.5352,
      "eval_samples_per_second": 12.431,
      "eval_steps_per_second": 1.59,
      "step": 800
    },
    {
      "epoch": 2.48,
      "eval_accuracy": 0.7508196721311475,
      "eval_f1": 0.6470554102382174,
      "eval_loss": 0.7644699215888977,
      "eval_precision": 0.6814040333813435,
      "eval_recall": 0.6477453030205323,
      "eval_runtime": 24.2035,
      "eval_samples_per_second": 12.601,
      "eval_steps_per_second": 1.611,
      "step": 850
    },
    {
      "epoch": 2.62,
      "eval_accuracy": 0.8,
      "eval_f1": 0.677888944499949,
      "eval_loss": 0.7002153396606445,
      "eval_precision": 0.7160759040458288,
      "eval_recall": 0.6559052650795769,
      "eval_runtime": 24.4891,
      "eval_samples_per_second": 12.455,
      "eval_steps_per_second": 1.593,
      "step": 900
    },
    {
      "epoch": 2.77,
      "eval_accuracy": 0.8098360655737705,
      "eval_f1": 0.6772457805812352,
      "eval_loss": 0.6718072891235352,
      "eval_precision": 0.6894471952583333,
      "eval_recall": 0.6719617414571543,
      "eval_runtime": 25.7261,
      "eval_samples_per_second": 11.856,
      "eval_steps_per_second": 1.516,
      "step": 950
    },
    {
      "epoch": 2.92,
      "grad_norm": 6.155330181121826,
      "learning_rate": 2.0845481049562683e-05,
      "loss": 0.6665,
      "step": 1000
    },
    {
      "epoch": 2.92,
      "eval_accuracy": 0.7868852459016393,
      "eval_f1": 0.6645166174577939,
      "eval_loss": 0.6745351552963257,
      "eval_precision": 0.6831230788775687,
      "eval_recall": 0.6611539278970471,
      "eval_runtime": 23.8162,
      "eval_samples_per_second": 12.806,
      "eval_steps_per_second": 1.638,
      "step": 1000
    },
    {
      "epoch": 3.06,
      "eval_accuracy": 0.8,
      "eval_f1": 0.6707482450301798,
      "eval_loss": 0.6912632584571838,
      "eval_precision": 0.7038907777116936,
      "eval_recall": 0.6636155184320323,
      "eval_runtime": 25.3425,
      "eval_samples_per_second": 12.035,
      "eval_steps_per_second": 1.539,
      "step": 1050
    },
    {
      "epoch": 3.21,
      "eval_accuracy": 0.7344262295081967,
      "eval_f1": 0.6124227128896254,
      "eval_loss": 0.839020848274231,
      "eval_precision": 0.6036382548711792,
      "eval_recall": 0.6331067003085352,
      "eval_runtime": 25.3402,
      "eval_samples_per_second": 12.036,
      "eval_steps_per_second": 1.539,
      "step": 1100
    },
    {
      "epoch": 3.35,
      "eval_accuracy": 0.7901639344262295,
      "eval_f1": 0.6632550969550687,
      "eval_loss": 0.7742147445678711,
      "eval_precision": 0.6782513190376637,
      "eval_recall": 0.6595217659437844,
      "eval_runtime": 24.3329,
      "eval_samples_per_second": 12.534,
      "eval_steps_per_second": 1.603,
      "step": 1150
    },
    {
      "epoch": 3.5,
      "eval_accuracy": 0.8131147540983606,
      "eval_f1": 0.6727344613312225,
      "eval_loss": 0.6753232479095459,
      "eval_precision": 0.6878592989192068,
      "eval_recall": 0.6654702186812278,
      "eval_runtime": 24.7418,
      "eval_samples_per_second": 12.327,
      "eval_steps_per_second": 1.576,
      "step": 1200
    },
    {
      "epoch": 3.64,
      "eval_accuracy": 0.760655737704918,
      "eval_f1": 0.6525488286572031,
      "eval_loss": 0.8542010188102722,
      "eval_precision": 0.6985802437047138,
      "eval_recall": 0.6503794246913512,
      "eval_runtime": 23.8367,
      "eval_samples_per_second": 12.795,
      "eval_steps_per_second": 1.636,
      "step": 1250
    },
    {
      "epoch": 3.79,
      "eval_accuracy": 0.8065573770491803,
      "eval_f1": 0.6740742285136359,
      "eval_loss": 0.7735392451286316,
      "eval_precision": 0.6767778649921506,
      "eval_recall": 0.676755828820049,
      "eval_runtime": 24.4555,
      "eval_samples_per_second": 12.472,
      "eval_steps_per_second": 1.595,
      "step": 1300
    },
    {
      "epoch": 3.94,
      "eval_accuracy": 0.8131147540983606,
      "eval_f1": 0.6832302669636724,
      "eval_loss": 0.6795472502708435,
      "eval_precision": 0.7133861439312567,
      "eval_recall": 0.6736156637532783,
      "eval_runtime": 25.0647,
      "eval_samples_per_second": 12.169,
      "eval_steps_per_second": 1.556,
      "step": 1350
    },
    {
      "epoch": 4.08,
      "eval_accuracy": 0.7967213114754098,
      "eval_f1": 0.6579730420842168,
      "eval_loss": 0.7954936623573303,
      "eval_precision": 0.6799340904944353,
      "eval_recall": 0.654890922780831,
      "eval_runtime": 25.0454,
      "eval_samples_per_second": 12.178,
      "eval_steps_per_second": 1.557,
      "step": 1400
    },
    {
      "epoch": 4.23,
      "eval_accuracy": 0.8229508196721311,
      "eval_f1": 0.6892471358587405,
      "eval_loss": 0.7999194264411926,
      "eval_precision": 0.7156453634629862,
      "eval_recall": 0.6809495048027159,
      "eval_runtime": 25.4511,
      "eval_samples_per_second": 11.984,
      "eval_steps_per_second": 1.532,
      "step": 1450
    },
    {
      "epoch": 4.37,
      "grad_norm": 22.507097244262695,
      "learning_rate": 6.268221574344024e-06,
      "loss": 0.3776,
      "step": 1500
    },
    {
      "epoch": 4.37,
      "eval_accuracy": 0.7934426229508197,
      "eval_f1": 0.6679894092876133,
      "eval_loss": 0.9446977376937866,
      "eval_precision": 0.686414129348912,
      "eval_recall": 0.6591507199305365,
      "eval_runtime": 25.6463,
      "eval_samples_per_second": 11.893,
      "eval_steps_per_second": 1.521,
      "step": 1500
    },
    {
      "epoch": 4.52,
      "eval_accuracy": 0.8065573770491803,
      "eval_f1": 0.6705779982361941,
      "eval_loss": 0.9103355407714844,
      "eval_precision": 0.6804748603360347,
      "eval_recall": 0.6682874981040119,
      "eval_runtime": 24.8615,
      "eval_samples_per_second": 12.268,
      "eval_steps_per_second": 1.569,
      "step": 1550
    },
    {
      "epoch": 4.66,
      "eval_accuracy": 0.7868852459016393,
      "eval_f1": 0.6581300883025021,
      "eval_loss": 0.9052944779396057,
      "eval_precision": 0.6659523069247808,
      "eval_recall": 0.6643693723968953,
      "eval_runtime": 24.6745,
      "eval_samples_per_second": 12.361,
      "eval_steps_per_second": 1.581,
      "step": 1600
    },
    {
      "epoch": 4.81,
      "eval_accuracy": 0.819672131147541,
      "eval_f1": 0.6860677167570124,
      "eval_loss": 0.85903000831604,
      "eval_precision": 0.7103231643472162,
      "eval_recall": 0.678005387179699,
      "eval_runtime": 25.3638,
      "eval_samples_per_second": 12.025,
      "eval_steps_per_second": 1.538,
      "step": 1650
    },
    {
      "epoch": 4.96,
      "eval_accuracy": 0.8229508196721311,
      "eval_f1": 0.6861351260633056,
      "eval_loss": 0.8571685552597046,
      "eval_precision": 0.7088281080916173,
      "eval_recall": 0.6787861482815611,
      "eval_runtime": 23.7505,
      "eval_samples_per_second": 12.842,
      "eval_steps_per_second": 1.642,
      "step": 1700
    }
  ],
  "logging_steps": 500,
  "max_steps": 1715,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.278165457700352e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}