trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9878260869565216,
  "eval_steps": 500,
  "global_step": 213,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06956521739130435,
      "grad_norm": 1.1811336278915405,
      "learning_rate": 4.99565044765711e-05,
      "loss": 0.7566,
      "num_input_tokens_seen": 137920,
      "step": 5
    },
    {
      "epoch": 0.1391304347826087,
      "grad_norm": 0.6466936469078064,
      "learning_rate": 4.978006327248537e-05,
      "loss": 0.6594,
      "num_input_tokens_seen": 271824,
      "step": 10
    },
    {
      "epoch": 0.20869565217391303,
      "grad_norm": 0.7790480852127075,
      "learning_rate": 4.946891632198452e-05,
      "loss": 0.6611,
      "num_input_tokens_seen": 411200,
      "step": 15
    },
    {
      "epoch": 0.2782608695652174,
      "grad_norm": 0.6762462258338928,
      "learning_rate": 4.9024755036496795e-05,
      "loss": 0.6368,
      "num_input_tokens_seen": 544816,
      "step": 20
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.5914238095283508,
      "learning_rate": 4.8449993900474187e-05,
      "loss": 0.6475,
      "num_input_tokens_seen": 682528,
      "step": 25
    },
    {
      "epoch": 0.41739130434782606,
      "grad_norm": 0.7006980776786804,
      "learning_rate": 4.774775734612604e-05,
      "loss": 0.6333,
      "num_input_tokens_seen": 813456,
      "step": 30
    },
    {
      "epoch": 0.48695652173913045,
      "grad_norm": 0.6066564321517944,
      "learning_rate": 4.6921862768838855e-05,
      "loss": 0.581,
      "num_input_tokens_seen": 949040,
      "step": 35
    },
    {
      "epoch": 0.5565217391304348,
      "grad_norm": 0.5768821835517883,
      "learning_rate": 4.597679977561122e-05,
      "loss": 0.5915,
      "num_input_tokens_seen": 1085984,
      "step": 40
    },
    {
      "epoch": 0.6260869565217392,
      "grad_norm": 0.7307961583137512,
      "learning_rate": 4.491770577931057e-05,
      "loss": 0.5869,
      "num_input_tokens_seen": 1217520,
      "step": 45
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.6931986808776855,
      "learning_rate": 4.375033807142267e-05,
      "loss": 0.5684,
      "num_input_tokens_seen": 1353968,
      "step": 50
    },
    {
      "epoch": 0.7652173913043478,
      "grad_norm": 0.5990815758705139,
      "learning_rate": 4.2481042525107854e-05,
      "loss": 0.5967,
      "num_input_tokens_seen": 1491728,
      "step": 55
    },
    {
      "epoch": 0.8347826086956521,
      "grad_norm": 0.5957231521606445,
      "learning_rate": 4.111671909869582e-05,
      "loss": 0.5831,
      "num_input_tokens_seen": 1625184,
      "step": 60
    },
    {
      "epoch": 0.9043478260869565,
      "grad_norm": 0.6507815718650818,
      "learning_rate": 3.9664784327143955e-05,
      "loss": 0.5782,
      "num_input_tokens_seen": 1758384,
      "step": 65
    },
    {
      "epoch": 0.9739130434782609,
      "grad_norm": 0.6880366802215576,
      "learning_rate": 3.813313100535747e-05,
      "loss": 0.5842,
      "num_input_tokens_seen": 1893088,
      "step": 70
    },
    {
      "epoch": 1.0556521739130436,
      "grad_norm": 0.6173655390739441,
      "learning_rate": 3.653008528253509e-05,
      "loss": 0.5629,
      "num_input_tokens_seen": 2051200,
      "step": 75
    },
    {
      "epoch": 1.1252173913043477,
      "grad_norm": 0.7451260685920715,
      "learning_rate": 3.486436140077764e-05,
      "loss": 0.5549,
      "num_input_tokens_seen": 2183136,
      "step": 80
    },
    {
      "epoch": 1.1947826086956521,
      "grad_norm": 0.6699721217155457,
      "learning_rate": 3.3145014324002944e-05,
      "loss": 0.5263,
      "num_input_tokens_seen": 2317616,
      "step": 85
    },
    {
      "epoch": 1.2643478260869565,
      "grad_norm": 0.6695335507392883,
      "learning_rate": 3.1381390514678696e-05,
      "loss": 0.505,
      "num_input_tokens_seen": 2457440,
      "step": 90
    },
    {
      "epoch": 1.333913043478261,
      "grad_norm": 0.7196683287620544,
      "learning_rate": 2.9583077125953716e-05,
      "loss": 0.555,
      "num_input_tokens_seen": 2592320,
      "step": 95
    },
    {
      "epoch": 1.4034782608695653,
      "grad_norm": 0.6481020450592041,
      "learning_rate": 2.775984988538175e-05,
      "loss": 0.5266,
      "num_input_tokens_seen": 2729856,
      "step": 100
    },
    {
      "epoch": 1.4730434782608697,
      "grad_norm": 0.8577863574028015,
      "learning_rate": 2.592161995354479e-05,
      "loss": 0.5364,
      "num_input_tokens_seen": 2866464,
      "step": 105
    },
    {
      "epoch": 1.542608695652174,
      "grad_norm": 0.7868128418922424,
      "learning_rate": 2.4078380046455222e-05,
      "loss": 0.5215,
      "num_input_tokens_seen": 3003392,
      "step": 110
    },
    {
      "epoch": 1.6121739130434782,
      "grad_norm": 0.8645579218864441,
      "learning_rate": 2.224015011461826e-05,
      "loss": 0.5354,
      "num_input_tokens_seen": 3141984,
      "step": 115
    },
    {
      "epoch": 1.6817391304347826,
      "grad_norm": 0.6744593381881714,
      "learning_rate": 2.0416922874046293e-05,
      "loss": 0.5098,
      "num_input_tokens_seen": 3281264,
      "step": 120
    },
    {
      "epoch": 1.7513043478260868,
      "grad_norm": 0.808380126953125,
      "learning_rate": 1.8618609485321313e-05,
      "loss": 0.4939,
      "num_input_tokens_seen": 3418208,
      "step": 125
    },
    {
      "epoch": 1.8208695652173912,
      "grad_norm": 0.7624004483222961,
      "learning_rate": 1.6854985675997066e-05,
      "loss": 0.4919,
      "num_input_tokens_seen": 3552624,
      "step": 130
    },
    {
      "epoch": 1.8904347826086956,
      "grad_norm": 0.8070163726806641,
      "learning_rate": 1.5135638599222368e-05,
      "loss": 0.5245,
      "num_input_tokens_seen": 3686752,
      "step": 135
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.9934524297714233,
      "learning_rate": 1.3469914717464916e-05,
      "loss": 0.5566,
      "num_input_tokens_seen": 3809216,
      "step": 140
    },
    {
      "epoch": 2.0417391304347827,
      "grad_norm": 0.855629563331604,
      "learning_rate": 1.1866868994642535e-05,
      "loss": 0.5672,
      "num_input_tokens_seen": 3971664,
      "step": 145
    },
    {
      "epoch": 2.111304347826087,
      "grad_norm": 0.8984305262565613,
      "learning_rate": 1.0335215672856046e-05,
      "loss": 0.5116,
      "num_input_tokens_seen": 4108736,
      "step": 150
    },
    {
      "epoch": 2.1808695652173915,
      "grad_norm": 0.9083377718925476,
      "learning_rate": 8.883280901304187e-06,
      "loss": 0.4855,
      "num_input_tokens_seen": 4246544,
      "step": 155
    },
    {
      "epoch": 2.2504347826086954,
      "grad_norm": 0.8181976079940796,
      "learning_rate": 7.518957474892149e-06,
      "loss": 0.4634,
      "num_input_tokens_seen": 4385600,
      "step": 160
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.8508033156394958,
      "learning_rate": 6.2496619285773356e-06,
      "loss": 0.4784,
      "num_input_tokens_seen": 4515424,
      "step": 165
    },
    {
      "epoch": 2.3895652173913042,
      "grad_norm": 0.813805103302002,
      "learning_rate": 5.082294220689435e-06,
      "loss": 0.4562,
      "num_input_tokens_seen": 4656832,
      "step": 170
    },
    {
      "epoch": 2.4591304347826086,
      "grad_norm": 0.8060736656188965,
      "learning_rate": 4.023200224388787e-06,
      "loss": 0.4635,
      "num_input_tokens_seen": 4787584,
      "step": 175
    },
    {
      "epoch": 2.528695652173913,
      "grad_norm": 0.9117311239242554,
      "learning_rate": 3.078137231161146e-06,
      "loss": 0.5231,
      "num_input_tokens_seen": 4920640,
      "step": 180
    },
    {
      "epoch": 2.5982608695652174,
      "grad_norm": 0.751384437084198,
      "learning_rate": 2.2522426538739566e-06,
      "loss": 0.4908,
      "num_input_tokens_seen": 5061360,
      "step": 185
    },
    {
      "epoch": 2.667826086956522,
      "grad_norm": 0.9320093989372253,
      "learning_rate": 1.5500060995258137e-06,
      "loss": 0.5252,
      "num_input_tokens_seen": 5192048,
      "step": 190
    },
    {
      "epoch": 2.737391304347826,
      "grad_norm": 0.7602404356002808,
      "learning_rate": 9.75244963503205e-07,
      "loss": 0.4728,
      "num_input_tokens_seen": 5317552,
      "step": 195
    },
    {
      "epoch": 2.8069565217391306,
      "grad_norm": 0.7692611813545227,
      "learning_rate": 5.310836780154899e-07,
      "loss": 0.4781,
      "num_input_tokens_seen": 5456864,
      "step": 200
    },
    {
      "epoch": 2.8765217391304345,
      "grad_norm": 0.9036366939544678,
      "learning_rate": 2.1993672751463579e-07,
      "loss": 0.5074,
      "num_input_tokens_seen": 5592320,
      "step": 205
    },
    {
      "epoch": 2.9460869565217394,
      "grad_norm": 0.8938913345336914,
      "learning_rate": 4.3495523428899174e-08,
      "loss": 0.454,
      "num_input_tokens_seen": 5728832,
      "step": 210
    },
    {
      "epoch": 2.9878260869565216,
      "num_input_tokens_seen": 5812176,
      "step": 213,
      "total_flos": 1.3025650857335194e+17,
      "train_loss": 0.5454816756673821,
      "train_runtime": 5250.0748,
      "train_samples_per_second": 0.657,
      "train_steps_per_second": 0.041
    }
  ],
  "logging_steps": 5,
  "max_steps": 213,
  "num_input_tokens_seen": 5812176,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3025650857335194e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}