last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00037837224261228196,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.7837224261228196e-05,
      "grad_norm": 14.805928230285645,
      "learning_rate": 2e-05,
      "loss": 4.2775,
      "step": 1
    },
    {
      "epoch": 3.7837224261228196e-05,
      "eval_loss": 1.0869630575180054,
      "eval_runtime": 2378.1464,
      "eval_samples_per_second": 2.34,
      "eval_steps_per_second": 2.34,
      "step": 1
    },
    {
      "epoch": 7.567444852245639e-05,
      "grad_norm": 14.053690910339355,
      "learning_rate": 4e-05,
      "loss": 4.198,
      "step": 2
    },
    {
      "epoch": 0.00011351167278368459,
      "grad_norm": 9.68853759765625,
      "learning_rate": 6e-05,
      "loss": 4.309,
      "step": 3
    },
    {
      "epoch": 0.00011351167278368459,
      "eval_loss": 0.9365310072898865,
      "eval_runtime": 2383.4538,
      "eval_samples_per_second": 2.334,
      "eval_steps_per_second": 2.334,
      "step": 3
    },
    {
      "epoch": 0.00015134889704491279,
      "grad_norm": 7.027285099029541,
      "learning_rate": 8e-05,
      "loss": 3.8788,
      "step": 4
    },
    {
      "epoch": 0.00018918612130614098,
      "grad_norm": 4.5182366371154785,
      "learning_rate": 0.0001,
      "loss": 3.2846,
      "step": 5
    },
    {
      "epoch": 0.00022702334556736918,
      "grad_norm": 3.581972122192383,
      "learning_rate": 0.00012,
      "loss": 3.2404,
      "step": 6
    },
    {
      "epoch": 0.00022702334556736918,
      "eval_loss": 0.7990255951881409,
      "eval_runtime": 2317.8914,
      "eval_samples_per_second": 2.4,
      "eval_steps_per_second": 2.4,
      "step": 6
    },
    {
      "epoch": 0.00026486056982859735,
      "grad_norm": 4.4711103439331055,
      "learning_rate": 0.00014,
      "loss": 3.3662,
      "step": 7
    },
    {
      "epoch": 0.00030269779408982557,
      "grad_norm": 3.5810935497283936,
      "learning_rate": 0.00016,
      "loss": 2.9538,
      "step": 8
    },
    {
      "epoch": 0.00034053501835105374,
      "grad_norm": 3.5936524868011475,
      "learning_rate": 0.00018,
      "loss": 2.434,
      "step": 9
    },
    {
      "epoch": 0.00034053501835105374,
      "eval_loss": 0.6861280202865601,
      "eval_runtime": 2598.99,
      "eval_samples_per_second": 2.141,
      "eval_steps_per_second": 2.141,
      "step": 9
    },
    {
      "epoch": 0.00037837224261228196,
      "grad_norm": 2.7024037837982178,
      "learning_rate": 0.0002,
      "loss": 2.68,
      "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": 7031294986813440.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}