last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00030419870259253346,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.0419870259253344e-05,
      "grad_norm": 1.5419278144836426,
      "learning_rate": 2e-05,
      "loss": 3.3333,
      "step": 1
    },
    {
      "epoch": 3.0419870259253344e-05,
      "eval_loss": 3.3719170093536377,
      "eval_runtime": 260.137,
      "eval_samples_per_second": 53.21,
      "eval_steps_per_second": 26.605,
      "step": 1
    },
    {
      "epoch": 6.083974051850669e-05,
      "grad_norm": 1.598204255104065,
      "learning_rate": 4e-05,
      "loss": 3.1842,
      "step": 2
    },
    {
      "epoch": 9.125961077776003e-05,
      "grad_norm": 1.5833808183670044,
      "learning_rate": 6e-05,
      "loss": 3.2607,
      "step": 3
    },
    {
      "epoch": 9.125961077776003e-05,
      "eval_loss": 3.3676836490631104,
      "eval_runtime": 259.0345,
      "eval_samples_per_second": 53.437,
      "eval_steps_per_second": 26.718,
      "step": 3
    },
    {
      "epoch": 0.00012167948103701338,
      "grad_norm": 1.505920171737671,
      "learning_rate": 8e-05,
      "loss": 3.2585,
      "step": 4
    },
    {
      "epoch": 0.00015209935129626673,
      "grad_norm": 1.5030511617660522,
      "learning_rate": 0.0001,
      "loss": 3.3801,
      "step": 5
    },
    {
      "epoch": 0.00018251922155552007,
      "grad_norm": 1.3640235662460327,
      "learning_rate": 0.00012,
      "loss": 3.3008,
      "step": 6
    },
    {
      "epoch": 0.00018251922155552007,
      "eval_loss": 3.328984498977661,
      "eval_runtime": 260.0701,
      "eval_samples_per_second": 53.224,
      "eval_steps_per_second": 26.612,
      "step": 6
    },
    {
      "epoch": 0.0002129390918147734,
      "grad_norm": 1.4955767393112183,
      "learning_rate": 0.00014,
      "loss": 3.8603,
      "step": 7
    },
    {
      "epoch": 0.00024335896207402675,
      "grad_norm": 1.2795424461364746,
      "learning_rate": 0.00016,
      "loss": 3.2918,
      "step": 8
    },
    {
      "epoch": 0.0002737788323332801,
      "grad_norm": 1.2046219110488892,
      "learning_rate": 0.00018,
      "loss": 3.0635,
      "step": 9
    },
    {
      "epoch": 0.0002737788323332801,
      "eval_loss": 3.28027081489563,
      "eval_runtime": 259.8702,
      "eval_samples_per_second": 53.265,
      "eval_steps_per_second": 26.633,
      "step": 9
    },
    {
      "epoch": 0.00030419870259253346,
      "grad_norm": 1.2982311248779297,
      "learning_rate": 0.0002,
      "loss": 3.3955,
      "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": 89038163804160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}