checkpoint-20/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.01999000499750125,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 3.3340553137590683,
      "learning_rate": 0.0,
      "loss": 11.0,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.398799707355898,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 10.125,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.3943029297945575,
      "learning_rate": 0.00011999999999999999,
      "loss": 10.1172,
      "step": 3
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.9959117709404242,
      "learning_rate": 0.00017999999999999998,
      "loss": 9.875,
      "step": 4
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8270696218303057,
      "learning_rate": 0.00023999999999999998,
      "loss": 9.6641,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.7854351602113614,
      "learning_rate": 0.0003,
      "loss": 9.4844,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.7194174424274788,
      "learning_rate": 0.00035999999999999997,
      "loss": 9.3281,
      "step": 7
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.463772638994466,
      "learning_rate": 0.00041999999999999996,
      "loss": 9.2109,
      "step": 8
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.439323678271545,
      "learning_rate": 0.00047999999999999996,
      "loss": 8.9453,
      "step": 9
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.2936126396494727,
      "learning_rate": 0.00054,
      "loss": 8.7109,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.0757761814549318,
      "learning_rate": 0.0005999986405514987,
      "loss": 8.4609,
      "step": 11
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9278570154341632,
      "learning_rate": 0.0005999945622196846,
      "loss": 8.2344,
      "step": 12
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8086775215724974,
      "learning_rate": 0.0005999877650456265,
      "loss": 8.125,
      "step": 13
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.7630413213242441,
      "learning_rate": 0.000599978249097772,
      "loss": 7.9766,
      "step": 14
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9172017565891333,
      "learning_rate": 0.0005999660144719463,
      "loss": 7.8555,
      "step": 15
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6610052304024877,
      "learning_rate": 0.0005999510612913519,
      "loss": 7.7734,
      "step": 16
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7091485456070775,
      "learning_rate": 0.0005999333897065673,
      "loss": 7.7148,
      "step": 17
    },
    {
      "epoch": 0.02,
      "grad_norm": 16.771353248766836,
      "learning_rate": 0.0005999129998955453,
      "loss": 8.5078,
      "step": 18
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.3123969082989795,
      "learning_rate": 0.0005998898920636111,
      "loss": 7.7539,
      "step": 19
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6992078172905232,
      "learning_rate": 0.00059986406644346,
      "loss": 7.75,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}