last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.011389521640091117,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0011389521640091116,
      "grad_norm": 1.0924690961837769,
      "learning_rate": 2e-05,
      "loss": 0.8415,
      "step": 1
    },
    {
      "epoch": 0.0011389521640091116,
      "eval_loss": 0.8290964365005493,
      "eval_runtime": 108.792,
      "eval_samples_per_second": 3.401,
      "eval_steps_per_second": 1.7,
      "step": 1
    },
    {
      "epoch": 0.002277904328018223,
      "grad_norm": 0.9600780606269836,
      "learning_rate": 4e-05,
      "loss": 0.8646,
      "step": 2
    },
    {
      "epoch": 0.003416856492027335,
      "grad_norm": 1.1602481603622437,
      "learning_rate": 6e-05,
      "loss": 0.879,
      "step": 3
    },
    {
      "epoch": 0.003416856492027335,
      "eval_loss": 0.8192018270492554,
      "eval_runtime": 113.1341,
      "eval_samples_per_second": 3.27,
      "eval_steps_per_second": 1.635,
      "step": 3
    },
    {
      "epoch": 0.004555808656036446,
      "grad_norm": 0.9277004599571228,
      "learning_rate": 8e-05,
      "loss": 0.7399,
      "step": 4
    },
    {
      "epoch": 0.0056947608200455585,
      "grad_norm": 0.9956090450286865,
      "learning_rate": 0.0001,
      "loss": 0.7967,
      "step": 5
    },
    {
      "epoch": 0.00683371298405467,
      "grad_norm": 0.8314662575721741,
      "learning_rate": 0.00012,
      "loss": 0.8103,
      "step": 6
    },
    {
      "epoch": 0.00683371298405467,
      "eval_loss": 0.7024092674255371,
      "eval_runtime": 111.3291,
      "eval_samples_per_second": 3.323,
      "eval_steps_per_second": 1.662,
      "step": 6
    },
    {
      "epoch": 0.007972665148063782,
      "grad_norm": 0.7975425124168396,
      "learning_rate": 0.00014,
      "loss": 0.5838,
      "step": 7
    },
    {
      "epoch": 0.009111617312072893,
      "grad_norm": 0.7116414904594421,
      "learning_rate": 0.00016,
      "loss": 0.533,
      "step": 8
    },
    {
      "epoch": 0.010250569476082005,
      "grad_norm": 0.8893793225288391,
      "learning_rate": 0.00018,
      "loss": 0.5525,
      "step": 9
    },
    {
      "epoch": 0.010250569476082005,
      "eval_loss": 0.5062820911407471,
      "eval_runtime": 111.9938,
      "eval_samples_per_second": 3.304,
      "eval_steps_per_second": 1.652,
      "step": 9
    },
    {
      "epoch": 0.011389521640091117,
      "grad_norm": 1.0999385118484497,
      "learning_rate": 0.0002,
      "loss": 0.509,
      "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": 2612537673646080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}