last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03991020204539786,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003991020204539786,
      "grad_norm": 8.886835098266602,
      "learning_rate": 1e-05,
      "loss": 3.8879,
      "step": 1
    },
    {
      "epoch": 0.003991020204539786,
      "eval_loss": 3.8941354751586914,
      "eval_runtime": 42.7144,
      "eval_samples_per_second": 9.88,
      "eval_steps_per_second": 4.94,
      "step": 1
    },
    {
      "epoch": 0.007982040409079572,
      "grad_norm": 8.932036399841309,
      "learning_rate": 2e-05,
      "loss": 3.6091,
      "step": 2
    },
    {
      "epoch": 0.011973060613619356,
      "grad_norm": 9.210203170776367,
      "learning_rate": 3e-05,
      "loss": 3.7088,
      "step": 3
    },
    {
      "epoch": 0.011973060613619356,
      "eval_loss": 3.6933300495147705,
      "eval_runtime": 3.2382,
      "eval_samples_per_second": 130.32,
      "eval_steps_per_second": 65.16,
      "step": 3
    },
    {
      "epoch": 0.015964080818159143,
      "grad_norm": 8.262584686279297,
      "learning_rate": 4e-05,
      "loss": 3.3967,
      "step": 4
    },
    {
      "epoch": 0.01995510102269893,
      "grad_norm": 8.927597999572754,
      "learning_rate": 5e-05,
      "loss": 3.5668,
      "step": 5
    },
    {
      "epoch": 0.02394612122723871,
      "grad_norm": 8.989409446716309,
      "learning_rate": 6e-05,
      "loss": 3.4426,
      "step": 6
    },
    {
      "epoch": 0.02394612122723871,
      "eval_loss": 3.1682167053222656,
      "eval_runtime": 3.3526,
      "eval_samples_per_second": 125.873,
      "eval_steps_per_second": 62.937,
      "step": 6
    },
    {
      "epoch": 0.027937141431778497,
      "grad_norm": 7.369055271148682,
      "learning_rate": 7e-05,
      "loss": 3.0186,
      "step": 7
    },
    {
      "epoch": 0.031928161636318286,
      "grad_norm": 6.372437953948975,
      "learning_rate": 8e-05,
      "loss": 3.0004,
      "step": 8
    },
    {
      "epoch": 0.03591918184085807,
      "grad_norm": 5.366674900054932,
      "learning_rate": 9e-05,
      "loss": 2.7329,
      "step": 9
    },
    {
      "epoch": 0.03591918184085807,
      "eval_loss": 2.4812095165252686,
      "eval_runtime": 3.4086,
      "eval_samples_per_second": 123.805,
      "eval_steps_per_second": 61.902,
      "step": 9
    },
    {
      "epoch": 0.03991020204539786,
      "grad_norm": 4.831432819366455,
      "learning_rate": 0.0001,
      "loss": 2.3456,
      "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": 356152655216640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}