last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.011604293588627792,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0011604293588627793,
      "grad_norm": 0.7944301962852478,
      "learning_rate": 2e-05,
      "loss": 0.7467,
      "step": 1
    },
    {
      "epoch": 0.0011604293588627793,
      "eval_loss": 0.8326185941696167,
      "eval_runtime": 71.8908,
      "eval_samples_per_second": 5.049,
      "eval_steps_per_second": 2.532,
      "step": 1
    },
    {
      "epoch": 0.0023208587177255585,
      "grad_norm": 0.3806622624397278,
      "learning_rate": 4e-05,
      "loss": 0.5817,
      "step": 2
    },
    {
      "epoch": 0.0034812880765883376,
      "grad_norm": 0.49611735343933105,
      "learning_rate": 6e-05,
      "loss": 0.7386,
      "step": 3
    },
    {
      "epoch": 0.0034812880765883376,
      "eval_loss": 0.8255924582481384,
      "eval_runtime": 72.0457,
      "eval_samples_per_second": 5.038,
      "eval_steps_per_second": 2.526,
      "step": 3
    },
    {
      "epoch": 0.004641717435451117,
      "grad_norm": 0.3392188847064972,
      "learning_rate": 8e-05,
      "loss": 0.4503,
      "step": 4
    },
    {
      "epoch": 0.005802146794313896,
      "grad_norm": 0.5582427382469177,
      "learning_rate": 0.0001,
      "loss": 0.7301,
      "step": 5
    },
    {
      "epoch": 0.006962576153176675,
      "grad_norm": 0.33241069316864014,
      "learning_rate": 0.00012,
      "loss": 0.3801,
      "step": 6
    },
    {
      "epoch": 0.006962576153176675,
      "eval_loss": 0.7233313322067261,
      "eval_runtime": 72.0034,
      "eval_samples_per_second": 5.041,
      "eval_steps_per_second": 2.528,
      "step": 6
    },
    {
      "epoch": 0.008123005512039455,
      "grad_norm": 0.4856230318546295,
      "learning_rate": 0.00014,
      "loss": 0.5106,
      "step": 7
    },
    {
      "epoch": 0.009283434870902234,
      "grad_norm": 0.32355302572250366,
      "learning_rate": 0.00016,
      "loss": 0.3177,
      "step": 8
    },
    {
      "epoch": 0.010443864229765013,
      "grad_norm": 0.7106059193611145,
      "learning_rate": 0.00018,
      "loss": 0.4037,
      "step": 9
    },
    {
      "epoch": 0.010443864229765013,
      "eval_loss": 0.5413673520088196,
      "eval_runtime": 71.8887,
      "eval_samples_per_second": 5.049,
      "eval_steps_per_second": 2.532,
      "step": 9
    },
    {
      "epoch": 0.011604293588627792,
      "grad_norm": 0.22967161238193512,
      "learning_rate": 0.0002,
      "loss": 0.2514,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2545835153817600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}