last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0036127167630057803,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000361271676300578,
      "grad_norm": 1.0353283882141113,
      "learning_rate": 2e-05,
      "loss": 2.255,
      "step": 1
    },
    {
      "epoch": 0.000361271676300578,
      "eval_loss": 2.2219531536102295,
      "eval_runtime": 46.7435,
      "eval_samples_per_second": 24.945,
      "eval_steps_per_second": 12.472,
      "step": 1
    },
    {
      "epoch": 0.000722543352601156,
      "grad_norm": 0.9947463870048523,
      "learning_rate": 4e-05,
      "loss": 1.7793,
      "step": 2
    },
    {
      "epoch": 0.001083815028901734,
      "grad_norm": 1.3340152502059937,
      "learning_rate": 6e-05,
      "loss": 2.687,
      "step": 3
    },
    {
      "epoch": 0.001083815028901734,
      "eval_loss": 2.2135438919067383,
      "eval_runtime": 46.8458,
      "eval_samples_per_second": 24.89,
      "eval_steps_per_second": 12.445,
      "step": 3
    },
    {
      "epoch": 0.001445086705202312,
      "grad_norm": 0.9923398494720459,
      "learning_rate": 8e-05,
      "loss": 2.0721,
      "step": 4
    },
    {
      "epoch": 0.0018063583815028901,
      "grad_norm": 1.245418667793274,
      "learning_rate": 0.0001,
      "loss": 2.4104,
      "step": 5
    },
    {
      "epoch": 0.002167630057803468,
      "grad_norm": 0.9161936640739441,
      "learning_rate": 0.00012,
      "loss": 2.0963,
      "step": 6
    },
    {
      "epoch": 0.002167630057803468,
      "eval_loss": 2.15303635597229,
      "eval_runtime": 46.8867,
      "eval_samples_per_second": 24.868,
      "eval_steps_per_second": 12.434,
      "step": 6
    },
    {
      "epoch": 0.0025289017341040463,
      "grad_norm": 0.7785300612449646,
      "learning_rate": 0.00014,
      "loss": 1.8938,
      "step": 7
    },
    {
      "epoch": 0.002890173410404624,
      "grad_norm": 0.898912787437439,
      "learning_rate": 0.00016,
      "loss": 1.835,
      "step": 8
    },
    {
      "epoch": 0.0032514450867052024,
      "grad_norm": 1.0745437145233154,
      "learning_rate": 0.00018,
      "loss": 1.5257,
      "step": 9
    },
    {
      "epoch": 0.0032514450867052024,
      "eval_loss": 2.0820212364196777,
      "eval_runtime": 46.8392,
      "eval_samples_per_second": 24.894,
      "eval_steps_per_second": 12.447,
      "step": 9
    },
    {
      "epoch": 0.0036127167630057803,
      "grad_norm": 0.9148722887039185,
      "learning_rate": 0.0002,
      "loss": 1.8243,
      "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": 2149977843302400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}