trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.2,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "grad_norm": 0.21985049545764923,
      "learning_rate": 0.0003304347826086957,
      "loss": 1.0688,
      "step": 20
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.21234580874443054,
      "learning_rate": 0.0003995221430894122,
      "loss": 0.932,
      "step": 40
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.23501691222190857,
      "learning_rate": 0.00039758476229578745,
      "loss": 0.9299,
      "step": 60
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.22043351829051971,
      "learning_rate": 0.0003941724426452488,
      "loss": 0.9042,
      "step": 80
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.23443147540092468,
      "learning_rate": 0.0003893106565618147,
      "loss": 0.9201,
      "step": 100
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.9053444862365723,
      "eval_runtime": 19.9264,
      "eval_samples_per_second": 25.092,
      "eval_steps_per_second": 3.162,
      "step": 100
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.24092139303684235,
      "learning_rate": 0.0003830356965061241,
      "loss": 0.9409,
      "step": 120
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.2834942638874054,
      "learning_rate": 0.0003753944040579839,
      "loss": 0.9091,
      "step": 140
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2784956991672516,
      "learning_rate": 0.00036644382025141837,
      "loss": 0.918,
      "step": 160
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.27023714780807495,
      "learning_rate": 0.0003562507597724135,
      "loss": 0.8686,
      "step": 180
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.22049479186534882,
      "learning_rate": 0.0003448913121979015,
      "loss": 0.8879,
      "step": 200
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.8908229470252991,
      "eval_runtime": 19.9283,
      "eval_samples_per_second": 25.09,
      "eval_steps_per_second": 3.161,
      "step": 200
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.2748544216156006,
      "learning_rate": 0.00033245027399915895,
      "loss": 0.8785,
      "step": 220
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.26322582364082336,
      "learning_rate": 0.0003190205155496219,
      "loss": 0.9239,
      "step": 240
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.22754715383052826,
      "learning_rate": 0.00030470228786230405,
      "loss": 0.864,
      "step": 260
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.23024863004684448,
      "learning_rate": 0.0002896024742319127,
      "loss": 0.7979,
      "step": 280
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.3036891222000122,
      "learning_rate": 0.0002738337923680367,
      "loss": 0.7781,
      "step": 300
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.8932064771652222,
      "eval_runtime": 19.9303,
      "eval_samples_per_second": 25.087,
      "eval_steps_per_second": 3.161,
      "step": 300
    }
  ],
  "logging_steps": 20,
  "max_steps": 750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.257643449975808e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}