last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.01931434089811685,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001931434089811685,
      "grad_norm": 0.28227633237838745,
      "learning_rate": 2e-05,
      "loss": 1.3072,
      "step": 1
    },
    {
      "epoch": 0.001931434089811685,
      "eval_loss": 1.3535726070404053,
      "eval_runtime": 81.7923,
      "eval_samples_per_second": 2.665,
      "eval_steps_per_second": 1.333,
      "step": 1
    },
    {
      "epoch": 0.00386286817962337,
      "grad_norm": 0.2996910512447357,
      "learning_rate": 4e-05,
      "loss": 1.531,
      "step": 2
    },
    {
      "epoch": 0.005794302269435055,
      "grad_norm": 0.2463427186012268,
      "learning_rate": 6e-05,
      "loss": 1.2405,
      "step": 3
    },
    {
      "epoch": 0.005794302269435055,
      "eval_loss": 1.3487764596939087,
      "eval_runtime": 83.012,
      "eval_samples_per_second": 2.626,
      "eval_steps_per_second": 1.313,
      "step": 3
    },
    {
      "epoch": 0.00772573635924674,
      "grad_norm": 0.4064790904521942,
      "learning_rate": 8e-05,
      "loss": 1.2321,
      "step": 4
    },
    {
      "epoch": 0.009657170449058426,
      "grad_norm": 0.2956888675689697,
      "learning_rate": 0.0001,
      "loss": 1.2542,
      "step": 5
    },
    {
      "epoch": 0.01158860453887011,
      "grad_norm": 0.2672729194164276,
      "learning_rate": 0.00012,
      "loss": 1.3084,
      "step": 6
    },
    {
      "epoch": 0.01158860453887011,
      "eval_loss": 1.3224433660507202,
      "eval_runtime": 83.1083,
      "eval_samples_per_second": 2.623,
      "eval_steps_per_second": 1.312,
      "step": 6
    },
    {
      "epoch": 0.013520038628681796,
      "grad_norm": 0.21218383312225342,
      "learning_rate": 0.00014,
      "loss": 1.4562,
      "step": 7
    },
    {
      "epoch": 0.01545147271849348,
      "grad_norm": 0.19916756451129913,
      "learning_rate": 0.00016,
      "loss": 1.1486,
      "step": 8
    },
    {
      "epoch": 0.017382906808305166,
      "grad_norm": 0.2508490979671478,
      "learning_rate": 0.00018,
      "loss": 1.2576,
      "step": 9
    },
    {
      "epoch": 0.017382906808305166,
      "eval_loss": 1.288987159729004,
      "eval_runtime": 83.0971,
      "eval_samples_per_second": 2.623,
      "eval_steps_per_second": 1.312,
      "step": 9
    },
    {
      "epoch": 0.01931434089811685,
      "grad_norm": 0.2675520181655884,
      "learning_rate": 0.0002,
      "loss": 1.1047,
      "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": 1.306916403806208e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}