last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.026917900403768506,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0026917900403768506,
      "grad_norm": 1.408136248588562,
      "learning_rate": 2e-05,
      "loss": 0.7399,
      "step": 1
    },
    {
      "epoch": 0.0026917900403768506,
      "eval_loss": 0.7382442355155945,
      "eval_runtime": 6.3386,
      "eval_samples_per_second": 12.463,
      "eval_steps_per_second": 12.463,
      "step": 1
    },
    {
      "epoch": 0.005383580080753701,
      "grad_norm": 1.402262568473816,
      "learning_rate": 4e-05,
      "loss": 0.7274,
      "step": 2
    },
    {
      "epoch": 0.008075370121130552,
      "grad_norm": 1.6652259826660156,
      "learning_rate": 6e-05,
      "loss": 0.7714,
      "step": 3
    },
    {
      "epoch": 0.008075370121130552,
      "eval_loss": 0.6933578848838806,
      "eval_runtime": 6.3351,
      "eval_samples_per_second": 12.47,
      "eval_steps_per_second": 12.47,
      "step": 3
    },
    {
      "epoch": 0.010767160161507403,
      "grad_norm": 1.429751992225647,
      "learning_rate": 8e-05,
      "loss": 0.7393,
      "step": 4
    },
    {
      "epoch": 0.013458950201884253,
      "grad_norm": 1.4376033544540405,
      "learning_rate": 0.0001,
      "loss": 0.7399,
      "step": 5
    },
    {
      "epoch": 0.016150740242261104,
      "grad_norm": 1.3412057161331177,
      "learning_rate": 0.00012,
      "loss": 0.6919,
      "step": 6
    },
    {
      "epoch": 0.016150740242261104,
      "eval_loss": 0.5569691061973572,
      "eval_runtime": 6.5483,
      "eval_samples_per_second": 12.064,
      "eval_steps_per_second": 12.064,
      "step": 6
    },
    {
      "epoch": 0.018842530282637954,
      "grad_norm": 1.139148473739624,
      "learning_rate": 0.00014,
      "loss": 0.4971,
      "step": 7
    },
    {
      "epoch": 0.021534320323014805,
      "grad_norm": 1.1586865186691284,
      "learning_rate": 0.00016,
      "loss": 0.4434,
      "step": 8
    },
    {
      "epoch": 0.024226110363391656,
      "grad_norm": 1.3152556419372559,
      "learning_rate": 0.00018,
      "loss": 0.4553,
      "step": 9
    },
    {
      "epoch": 0.024226110363391656,
      "eval_loss": 0.4071764349937439,
      "eval_runtime": 6.4684,
      "eval_samples_per_second": 12.213,
      "eval_steps_per_second": 12.213,
      "step": 9
    },
    {
      "epoch": 0.026917900403768506,
      "grad_norm": 1.2165873050689697,
      "learning_rate": 0.0002,
      "loss": 0.4257,
      "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": 2358194416386048.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}