last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008424599831508003,
  "eval_steps": 2,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008424599831508003,
      "grad_norm": 16.380062103271484,
      "learning_rate": 2e-05,
      "loss": 20.7642,
      "step": 1
    },
    {
      "epoch": 0.0008424599831508003,
      "eval_loss": 5.003799915313721,
      "eval_runtime": 86.8933,
      "eval_samples_per_second": 5.754,
      "eval_steps_per_second": 2.877,
      "step": 1
    },
    {
      "epoch": 0.0016849199663016006,
      "grad_norm": 17.466218948364258,
      "learning_rate": 4e-05,
      "loss": 18.4442,
      "step": 2
    },
    {
      "epoch": 0.0016849199663016006,
      "eval_loss": 4.992175579071045,
      "eval_runtime": 86.961,
      "eval_samples_per_second": 5.75,
      "eval_steps_per_second": 2.875,
      "step": 2
    },
    {
      "epoch": 0.002527379949452401,
      "grad_norm": 13.229546546936035,
      "learning_rate": 6e-05,
      "loss": 21.6802,
      "step": 3
    },
    {
      "epoch": 0.003369839932603201,
      "grad_norm": 22.8873348236084,
      "learning_rate": 8e-05,
      "loss": 18.9218,
      "step": 4
    },
    {
      "epoch": 0.003369839932603201,
      "eval_loss": 4.732216835021973,
      "eval_runtime": 87.2134,
      "eval_samples_per_second": 5.733,
      "eval_steps_per_second": 2.867,
      "step": 4
    },
    {
      "epoch": 0.004212299915754001,
      "grad_norm": 26.322071075439453,
      "learning_rate": 0.0001,
      "loss": 17.8213,
      "step": 5
    },
    {
      "epoch": 0.005054759898904802,
      "grad_norm": 29.4674072265625,
      "learning_rate": 0.00012,
      "loss": 17.2315,
      "step": 6
    },
    {
      "epoch": 0.005054759898904802,
      "eval_loss": 3.23167085647583,
      "eval_runtime": 87.097,
      "eval_samples_per_second": 5.741,
      "eval_steps_per_second": 2.87,
      "step": 6
    },
    {
      "epoch": 0.005897219882055603,
      "grad_norm": 25.561017990112305,
      "learning_rate": 0.00014,
      "loss": 12.8493,
      "step": 7
    },
    {
      "epoch": 0.006739679865206402,
      "grad_norm": 25.24022102355957,
      "learning_rate": 0.00016,
      "loss": 9.74,
      "step": 8
    },
    {
      "epoch": 0.006739679865206402,
      "eval_loss": 1.5246657133102417,
      "eval_runtime": 87.3812,
      "eval_samples_per_second": 5.722,
      "eval_steps_per_second": 2.861,
      "step": 8
    },
    {
      "epoch": 0.007582139848357203,
      "grad_norm": 19.95369529724121,
      "learning_rate": 0.00018,
      "loss": 5.6036,
      "step": 9
    },
    {
      "epoch": 0.008424599831508003,
      "grad_norm": 17.343950271606445,
      "learning_rate": 0.0002,
      "loss": 3.5369,
      "step": 10
    },
    {
      "epoch": 0.008424599831508003,
      "eval_loss": 0.4855375289916992,
      "eval_runtime": 86.9048,
      "eval_samples_per_second": 5.753,
      "eval_steps_per_second": 2.877,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6347789697024000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}