last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08620689655172414,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008620689655172414,
      "grad_norm": 1.8437737226486206,
      "learning_rate": 2e-05,
      "loss": 3.8128,
      "step": 1
    },
    {
      "epoch": 0.008620689655172414,
      "eval_loss": 4.169641017913818,
      "eval_runtime": 9.4681,
      "eval_samples_per_second": 20.596,
      "eval_steps_per_second": 2.64,
      "step": 1
    },
    {
      "epoch": 0.017241379310344827,
      "grad_norm": 1.5073466300964355,
      "learning_rate": 4e-05,
      "loss": 4.0167,
      "step": 2
    },
    {
      "epoch": 0.02586206896551724,
      "grad_norm": 1.684279203414917,
      "learning_rate": 6e-05,
      "loss": 4.0414,
      "step": 3
    },
    {
      "epoch": 0.02586206896551724,
      "eval_loss": 4.129457473754883,
      "eval_runtime": 9.4929,
      "eval_samples_per_second": 20.542,
      "eval_steps_per_second": 2.634,
      "step": 3
    },
    {
      "epoch": 0.034482758620689655,
      "grad_norm": 1.8984614610671997,
      "learning_rate": 8e-05,
      "loss": 4.18,
      "step": 4
    },
    {
      "epoch": 0.04310344827586207,
      "grad_norm": 1.8628818988800049,
      "learning_rate": 0.0001,
      "loss": 4.0706,
      "step": 5
    },
    {
      "epoch": 0.05172413793103448,
      "grad_norm": 2.0986077785491943,
      "learning_rate": 0.00012,
      "loss": 3.887,
      "step": 6
    },
    {
      "epoch": 0.05172413793103448,
      "eval_loss": 3.8300561904907227,
      "eval_runtime": 9.5102,
      "eval_samples_per_second": 20.504,
      "eval_steps_per_second": 2.629,
      "step": 6
    },
    {
      "epoch": 0.0603448275862069,
      "grad_norm": 2.188512086868286,
      "learning_rate": 0.00014,
      "loss": 3.6447,
      "step": 7
    },
    {
      "epoch": 0.06896551724137931,
      "grad_norm": 1.5209987163543701,
      "learning_rate": 0.00016,
      "loss": 3.8437,
      "step": 8
    },
    {
      "epoch": 0.07758620689655173,
      "grad_norm": 0.9197730422019958,
      "learning_rate": 0.00018,
      "loss": 3.2178,
      "step": 9
    },
    {
      "epoch": 0.07758620689655173,
      "eval_loss": 3.457179069519043,
      "eval_runtime": 9.5467,
      "eval_samples_per_second": 20.426,
      "eval_steps_per_second": 2.619,
      "step": 9
    },
    {
      "epoch": 0.08620689655172414,
      "grad_norm": 0.9381762742996216,
      "learning_rate": 0.0002,
      "loss": 3.5532,
      "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": 5.227665615224832e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}