trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.971014492753623,
  "eval_steps": 500,
  "global_step": 102,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2898550724637681,
      "grad_norm": 13.671546869740457,
      "learning_rate": 5e-06,
      "loss": 0.9276,
      "step": 10
    },
    {
      "epoch": 0.5797101449275363,
      "grad_norm": 9.031033261420918,
      "learning_rate": 5e-06,
      "loss": 0.8394,
      "step": 20
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 1.1037926170894152,
      "learning_rate": 5e-06,
      "loss": 0.7948,
      "step": 30
    },
    {
      "epoch": 0.9855072463768116,
      "eval_loss": 0.7815124988555908,
      "eval_runtime": 24.0599,
      "eval_samples_per_second": 38.654,
      "eval_steps_per_second": 0.623,
      "step": 34
    },
    {
      "epoch": 1.1666666666666667,
      "grad_norm": 1.2717518900808908,
      "learning_rate": 5e-06,
      "loss": 0.8103,
      "step": 40
    },
    {
      "epoch": 1.4565217391304348,
      "grad_norm": 1.9531380513897942,
      "learning_rate": 5e-06,
      "loss": 0.7223,
      "step": 50
    },
    {
      "epoch": 1.7463768115942029,
      "grad_norm": 0.8944216418486982,
      "learning_rate": 5e-06,
      "loss": 0.7137,
      "step": 60
    },
    {
      "epoch": 1.9782608695652173,
      "eval_loss": 0.7427083253860474,
      "eval_runtime": 23.9164,
      "eval_samples_per_second": 38.885,
      "eval_steps_per_second": 0.627,
      "step": 68
    },
    {
      "epoch": 2.0434782608695654,
      "grad_norm": 1.1873450717773506,
      "learning_rate": 5e-06,
      "loss": 0.751,
      "step": 70
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.9502277849330181,
      "learning_rate": 5e-06,
      "loss": 0.6497,
      "step": 80
    },
    {
      "epoch": 2.6231884057971016,
      "grad_norm": 0.9134400285354503,
      "learning_rate": 5e-06,
      "loss": 0.6473,
      "step": 90
    },
    {
      "epoch": 2.9130434782608696,
      "grad_norm": 0.7935078517776749,
      "learning_rate": 5e-06,
      "loss": 0.6436,
      "step": 100
    },
    {
      "epoch": 2.971014492753623,
      "eval_loss": 0.7359128594398499,
      "eval_runtime": 23.1152,
      "eval_samples_per_second": 40.233,
      "eval_steps_per_second": 0.649,
      "step": 102
    },
    {
      "epoch": 2.971014492753623,
      "step": 102,
      "total_flos": 170644419379200.0,
      "train_loss": 0.7483346707680646,
      "train_runtime": 3633.2075,
      "train_samples_per_second": 14.581,
      "train_steps_per_second": 0.028
    }
  ],
  "logging_steps": 10,
  "max_steps": 102,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 170644419379200.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}