last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.017543859649122806,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017543859649122807,
      "grad_norm": 1038.8984375,
      "learning_rate": 2e-05,
      "loss": 12.4343,
      "step": 1
    },
    {
      "epoch": 0.0017543859649122807,
      "eval_loss": 2.4816813468933105,
      "eval_runtime": 2.8231,
      "eval_samples_per_second": 85.013,
      "eval_steps_per_second": 42.507,
      "step": 1
    },
    {
      "epoch": 0.0035087719298245615,
      "grad_norm": 1177.747802734375,
      "learning_rate": 4e-05,
      "loss": 10.0677,
      "step": 2
    },
    {
      "epoch": 0.005263157894736842,
      "grad_norm": 989.3358764648438,
      "learning_rate": 6e-05,
      "loss": 6.45,
      "step": 3
    },
    {
      "epoch": 0.005263157894736842,
      "eval_loss": 2.4625728130340576,
      "eval_runtime": 1.8907,
      "eval_samples_per_second": 126.939,
      "eval_steps_per_second": 63.47,
      "step": 3
    },
    {
      "epoch": 0.007017543859649123,
      "grad_norm": 865.8417358398438,
      "learning_rate": 8e-05,
      "loss": 10.3514,
      "step": 4
    },
    {
      "epoch": 0.008771929824561403,
      "grad_norm": 982.4260864257812,
      "learning_rate": 0.0001,
      "loss": 12.5797,
      "step": 5
    },
    {
      "epoch": 0.010526315789473684,
      "grad_norm": 1129.5501708984375,
      "learning_rate": 0.00012,
      "loss": 9.1177,
      "step": 6
    },
    {
      "epoch": 0.010526315789473684,
      "eval_loss": 2.460749387741089,
      "eval_runtime": 1.9024,
      "eval_samples_per_second": 126.158,
      "eval_steps_per_second": 63.079,
      "step": 6
    },
    {
      "epoch": 0.012280701754385965,
      "grad_norm": 631.5029907226562,
      "learning_rate": 0.00014,
      "loss": 9.8423,
      "step": 7
    },
    {
      "epoch": 0.014035087719298246,
      "grad_norm": 993.0007934570312,
      "learning_rate": 0.00016,
      "loss": 7.2848,
      "step": 8
    },
    {
      "epoch": 0.015789473684210527,
      "grad_norm": 1311.847412109375,
      "learning_rate": 0.00018,
      "loss": 14.4826,
      "step": 9
    },
    {
      "epoch": 0.015789473684210527,
      "eval_loss": 2.4422192573547363,
      "eval_runtime": 1.8963,
      "eval_samples_per_second": 126.561,
      "eval_steps_per_second": 63.281,
      "step": 9
    },
    {
      "epoch": 0.017543859649122806,
      "grad_norm": 1083.43798828125,
      "learning_rate": 0.0002,
      "loss": 9.7772,
      "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": 48333340016640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}