moe_shakespeare15M/checkpoint-400/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 333.0,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 17.0,
      "grad_norm": 6.09660530090332,
      "learning_rate": 5e-06,
      "loss": 2.5103,
      "step": 20
    },
    {
      "epoch": 33.0,
      "grad_norm": 1.7365801334381104,
      "learning_rate": 1e-05,
      "loss": 2.1374,
      "step": 40
    },
    {
      "epoch": 50.0,
      "grad_norm": 1.1849949359893799,
      "learning_rate": 1.5e-05,
      "loss": 2.0202,
      "step": 60
    },
    {
      "epoch": 67.0,
      "grad_norm": 0.8566276431083679,
      "learning_rate": 2e-05,
      "loss": 1.8209,
      "step": 80
    },
    {
      "epoch": 83.0,
      "grad_norm": 0.733796238899231,
      "learning_rate": 2.5e-05,
      "loss": 1.5589,
      "step": 100
    },
    {
      "epoch": 100.0,
      "grad_norm": 1.1494934558868408,
      "learning_rate": 2.375e-05,
      "loss": 1.4981,
      "step": 120
    },
    {
      "epoch": 117.0,
      "grad_norm": 1.0989091396331787,
      "learning_rate": 2.25e-05,
      "loss": 1.3532,
      "step": 140
    },
    {
      "epoch": 133.0,
      "grad_norm": 1.2396273612976074,
      "learning_rate": 2.125e-05,
      "loss": 1.1437,
      "step": 160
    },
    {
      "epoch": 150.0,
      "grad_norm": 1.9112813472747803,
      "learning_rate": 2e-05,
      "loss": 1.0889,
      "step": 180
    },
    {
      "epoch": 167.0,
      "grad_norm": 1.9788281917572021,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.9825,
      "step": 200
    },
    {
      "epoch": 183.0,
      "grad_norm": 2.772813081741333,
      "learning_rate": 1.75e-05,
      "loss": 0.832,
      "step": 220
    },
    {
      "epoch": 200.0,
      "grad_norm": 2.384052038192749,
      "learning_rate": 1.6250000000000002e-05,
      "loss": 0.7939,
      "step": 240
    },
    {
      "epoch": 217.0,
      "grad_norm": 2.423220634460449,
      "learning_rate": 1.5e-05,
      "loss": 0.7084,
      "step": 260
    },
    {
      "epoch": 233.0,
      "grad_norm": 2.2296969890594482,
      "learning_rate": 1.3750000000000002e-05,
      "loss": 0.5999,
      "step": 280
    },
    {
      "epoch": 250.0,
      "grad_norm": 3.0638062953948975,
      "learning_rate": 1.25e-05,
      "loss": 0.5756,
      "step": 300
    },
    {
      "epoch": 267.0,
      "grad_norm": 3.0690948963165283,
      "learning_rate": 1.125e-05,
      "loss": 0.5252,
      "step": 320
    },
    {
      "epoch": 283.0,
      "grad_norm": 2.1726977825164795,
      "learning_rate": 1e-05,
      "loss": 0.4515,
      "step": 340
    },
    {
      "epoch": 300.0,
      "grad_norm": 2.711531639099121,
      "learning_rate": 8.75e-06,
      "loss": 0.4445,
      "step": 360
    },
    {
      "epoch": 317.0,
      "grad_norm": 2.583651542663574,
      "learning_rate": 7.5e-06,
      "loss": 0.4179,
      "step": 380
    },
    {
      "epoch": 333.0,
      "grad_norm": 2.864351511001587,
      "learning_rate": 6.25e-06,
      "loss": 0.363,
      "step": 400
    }
  ],
  "logging_steps": 20,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 500,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1439972197574400.0,
  "train_batch_size": 50,
  "trial_name": null,
  "trial_params": null
}