last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0072,
  "eval_steps": 500,
  "global_step": 360,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 79.12922668457031,
      "learning_rate": 1e-06,
      "loss": 4.6771,
      "step": 20
    },
    {
      "epoch": 0.0,
      "grad_norm": 239.7991943359375,
      "learning_rate": 2e-06,
      "loss": 4.5023,
      "step": 40
    },
    {
      "epoch": 0.0,
      "grad_norm": 77.5164794921875,
      "learning_rate": 3e-06,
      "loss": 4.8171,
      "step": 60
    },
    {
      "epoch": 0.0,
      "grad_norm": 207.87796020507812,
      "learning_rate": 4e-06,
      "loss": 4.6765,
      "step": 80
    },
    {
      "epoch": 0.0,
      "grad_norm": 147.12461853027344,
      "learning_rate": 4.9999999999999996e-06,
      "loss": 4.7237,
      "step": 100
    },
    {
      "epoch": 0.0,
      "grad_norm": 115.61023712158203,
      "learning_rate": 6e-06,
      "loss": 4.4426,
      "step": 120
    },
    {
      "epoch": 0.0,
      "grad_norm": 43.68627166748047,
      "learning_rate": 7e-06,
      "loss": 5.0142,
      "step": 140
    },
    {
      "epoch": 0.0,
      "grad_norm": 37.58155059814453,
      "learning_rate": 8e-06,
      "loss": 4.8572,
      "step": 160
    },
    {
      "epoch": 0.0,
      "grad_norm": 86.82076263427734,
      "learning_rate": 9e-06,
      "loss": 4.6848,
      "step": 180
    },
    {
      "epoch": 0.0,
      "grad_norm": 289.0603332519531,
      "learning_rate": 9.999999999999999e-06,
      "loss": 4.6474,
      "step": 200
    },
    {
      "epoch": 0.0,
      "grad_norm": 69.93185424804688,
      "learning_rate": 1.1e-05,
      "loss": 4.9724,
      "step": 220
    },
    {
      "epoch": 0.0,
      "grad_norm": 98.215087890625,
      "learning_rate": 1.2e-05,
      "loss": 4.4447,
      "step": 240
    },
    {
      "epoch": 0.01,
      "grad_norm": 92.3516845703125,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 4.9076,
      "step": 260
    },
    {
      "epoch": 0.01,
      "grad_norm": 150.6816864013672,
      "learning_rate": 1.4e-05,
      "loss": 4.8402,
      "step": 280
    },
    {
      "epoch": 0.01,
      "grad_norm": 162.9401397705078,
      "learning_rate": 1.5e-05,
      "loss": 4.9279,
      "step": 300
    },
    {
      "epoch": 0.01,
      "grad_norm": 38.77900695800781,
      "learning_rate": 1.6e-05,
      "loss": 4.5021,
      "step": 320
    },
    {
      "epoch": 0.01,
      "grad_norm": 91.0234146118164,
      "learning_rate": 1.7e-05,
      "loss": 4.5546,
      "step": 340
    },
    {
      "epoch": 0.01,
      "grad_norm": 78.24981689453125,
      "learning_rate": 1.8e-05,
      "loss": 4.3795,
      "step": 360
    }
  ],
  "logging_steps": 20,
  "max_steps": 20000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 20,
  "total_flos": 127576999796736.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}