trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.3823704586063132,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11911852293031566,
      "grad_norm": 0.3590008318424225,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.7627,
      "step": 100
    },
    {
      "epoch": 0.23823704586063132,
      "grad_norm": 0.5137978196144104,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6067,
      "step": 200
    },
    {
      "epoch": 0.357355568790947,
      "grad_norm": 0.46571800112724304,
      "learning_rate": 5e-05,
      "loss": 0.5131,
      "step": 300
    },
    {
      "epoch": 0.47647409172126265,
      "grad_norm": 0.48380178213119507,
      "learning_rate": 4.957432749209755e-05,
      "loss": 0.4791,
      "step": 400
    },
    {
      "epoch": 0.5955926146515783,
      "grad_norm": 0.5578784942626953,
      "learning_rate": 4.8311805735108894e-05,
      "loss": 0.4766,
      "step": 500
    },
    {
      "epoch": 0.714711137581894,
      "grad_norm": 0.4985765516757965,
      "learning_rate": 4.625542839324036e-05,
      "loss": 0.4803,
      "step": 600
    },
    {
      "epoch": 0.8338296605122096,
      "grad_norm": 0.5892139673233032,
      "learning_rate": 4.347522293051648e-05,
      "loss": 0.4718,
      "step": 700
    },
    {
      "epoch": 0.9529481834425253,
      "grad_norm": 0.5380077362060547,
      "learning_rate": 4.0065865909481417e-05,
      "loss": 0.4626,
      "step": 800
    },
    {
      "epoch": 1.072066706372841,
      "grad_norm": 0.576122522354126,
      "learning_rate": 3.6143458894413465e-05,
      "loss": 0.4564,
      "step": 900
    },
    {
      "epoch": 1.1911852293031566,
      "grad_norm": 0.543662428855896,
      "learning_rate": 3.1841574751802076e-05,
      "loss": 0.4526,
      "step": 1000
    },
    {
      "epoch": 1.3103037522334722,
      "grad_norm": 0.6222002506256104,
      "learning_rate": 2.7306708986582553e-05,
      "loss": 0.4626,
      "step": 1100
    },
    {
      "epoch": 1.429422275163788,
      "grad_norm": 0.5597625374794006,
      "learning_rate": 2.2693291013417453e-05,
      "loss": 0.4581,
      "step": 1200
    },
    {
      "epoch": 1.5485407980941037,
      "grad_norm": 0.8017860651016235,
      "learning_rate": 1.815842524819793e-05,
      "loss": 0.4646,
      "step": 1300
    },
    {
      "epoch": 1.6676593210244193,
      "grad_norm": 0.696183443069458,
      "learning_rate": 1.3856541105586545e-05,
      "loss": 0.4548,
      "step": 1400
    },
    {
      "epoch": 1.7867778439547348,
      "grad_norm": 0.6249508261680603,
      "learning_rate": 9.934134090518593e-06,
      "loss": 0.4668,
      "step": 1500
    },
    {
      "epoch": 1.9058963668850506,
      "grad_norm": 0.679349422454834,
      "learning_rate": 6.524777069483526e-06,
      "loss": 0.4519,
      "step": 1600
    },
    {
      "epoch": 2.0250148898153664,
      "grad_norm": 0.6183383464813232,
      "learning_rate": 3.7445716067596503e-06,
      "loss": 0.4554,
      "step": 1700
    },
    {
      "epoch": 2.144133412745682,
      "grad_norm": 0.7539538145065308,
      "learning_rate": 1.6881942648911076e-06,
      "loss": 0.4568,
      "step": 1800
    },
    {
      "epoch": 2.2632519356759975,
      "grad_norm": 0.7621870636940002,
      "learning_rate": 4.256725079024554e-07,
      "loss": 0.4577,
      "step": 1900
    },
    {
      "epoch": 2.3823704586063132,
      "grad_norm": 0.45056673884391785,
      "learning_rate": 0.0,
      "loss": 0.4445,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 2000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.15307218125783e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}