last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0321285140562249,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00321285140562249,
      "grad_norm": 10.668553352355957,
      "learning_rate": 2e-05,
      "loss": 5.299,
      "step": 1
    },
    {
      "epoch": 0.00321285140562249,
      "eval_loss": 5.279109477996826,
      "eval_runtime": 6.1635,
      "eval_samples_per_second": 85.016,
      "eval_steps_per_second": 10.708,
      "step": 1
    },
    {
      "epoch": 0.00642570281124498,
      "grad_norm": 10.448212623596191,
      "learning_rate": 4e-05,
      "loss": 4.6254,
      "step": 2
    },
    {
      "epoch": 0.00963855421686747,
      "grad_norm": 10.46432876586914,
      "learning_rate": 6e-05,
      "loss": 5.0187,
      "step": 3
    },
    {
      "epoch": 0.00963855421686747,
      "eval_loss": 4.949011325836182,
      "eval_runtime": 6.217,
      "eval_samples_per_second": 84.285,
      "eval_steps_per_second": 10.616,
      "step": 3
    },
    {
      "epoch": 0.01285140562248996,
      "grad_norm": 9.103033065795898,
      "learning_rate": 8e-05,
      "loss": 3.8705,
      "step": 4
    },
    {
      "epoch": 0.01606425702811245,
      "grad_norm": 9.697961807250977,
      "learning_rate": 0.0001,
      "loss": 4.6658,
      "step": 5
    },
    {
      "epoch": 0.01927710843373494,
      "grad_norm": 6.769638538360596,
      "learning_rate": 0.00012,
      "loss": 3.7274,
      "step": 6
    },
    {
      "epoch": 0.01927710843373494,
      "eval_loss": 3.3475189208984375,
      "eval_runtime": 6.2189,
      "eval_samples_per_second": 84.26,
      "eval_steps_per_second": 10.613,
      "step": 6
    },
    {
      "epoch": 0.02248995983935743,
      "grad_norm": 7.49973201751709,
      "learning_rate": 0.00014,
      "loss": 2.8421,
      "step": 7
    },
    {
      "epoch": 0.02570281124497992,
      "grad_norm": 7.562386512756348,
      "learning_rate": 0.00016,
      "loss": 2.1444,
      "step": 8
    },
    {
      "epoch": 0.02891566265060241,
      "grad_norm": 6.556207180023193,
      "learning_rate": 0.00018,
      "loss": 1.8961,
      "step": 9
    },
    {
      "epoch": 0.02891566265060241,
      "eval_loss": 2.141901731491089,
      "eval_runtime": 6.22,
      "eval_samples_per_second": 84.244,
      "eval_steps_per_second": 10.611,
      "step": 9
    },
    {
      "epoch": 0.0321285140562249,
      "grad_norm": 13.10632610321045,
      "learning_rate": 0.0002,
      "loss": 1.7408,
      "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": 8265470012555264.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}