trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 34.78260869565217,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 0.1043701171875,
      "learning_rate": 0.00019876883405951377,
      "loss": 0.5291,
      "step": 10
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.082275390625,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.509,
      "step": 20
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 0.145751953125,
      "learning_rate": 0.0001891006524188368,
      "loss": 0.4995,
      "step": 30
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.1356201171875,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.4853,
      "step": 40
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 0.1287841796875,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.4699,
      "step": 50
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.1466064453125,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.456,
      "step": 60
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 0.1771240234375,
      "learning_rate": 0.00014539904997395468,
      "loss": 0.4446,
      "step": 70
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.21875,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.4325,
      "step": 80
    },
    {
      "epoch": 15.652173913043478,
      "grad_norm": 0.25732421875,
      "learning_rate": 0.0001156434465040231,
      "loss": 0.4166,
      "step": 90
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 0.25830078125,
      "learning_rate": 0.0001,
      "loss": 0.4079,
      "step": 100
    },
    {
      "epoch": 19.130434782608695,
      "grad_norm": 0.34033203125,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.3981,
      "step": 110
    },
    {
      "epoch": 20.869565217391305,
      "grad_norm": 0.28369140625,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.3853,
      "step": 120
    },
    {
      "epoch": 22.608695652173914,
      "grad_norm": 0.27685546875,
      "learning_rate": 5.4600950026045326e-05,
      "loss": 0.3754,
      "step": 130
    },
    {
      "epoch": 24.347826086956523,
      "grad_norm": 0.297607421875,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.368,
      "step": 140
    },
    {
      "epoch": 26.08695652173913,
      "grad_norm": 0.398681640625,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 0.3674,
      "step": 150
    },
    {
      "epoch": 27.82608695652174,
      "grad_norm": 0.318115234375,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.3594,
      "step": 160
    },
    {
      "epoch": 29.565217391304348,
      "grad_norm": 0.3125,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.3593,
      "step": 170
    },
    {
      "epoch": 31.304347826086957,
      "grad_norm": 0.30126953125,
      "learning_rate": 4.8943483704846475e-06,
      "loss": 0.355,
      "step": 180
    },
    {
      "epoch": 33.04347826086956,
      "grad_norm": 0.401123046875,
      "learning_rate": 1.231165940486234e-06,
      "loss": 0.351,
      "step": 190
    },
    {
      "epoch": 34.78260869565217,
      "grad_norm": 0.295166015625,
      "learning_rate": 0.0,
      "loss": 0.3531,
      "step": 200
    },
    {
      "epoch": 34.78260869565217,
      "step": 200,
      "total_flos": 3.946780925991322e+16,
      "train_loss": 0.4161242663860321,
      "train_runtime": 575.6907,
      "train_samples_per_second": 11.117,
      "train_steps_per_second": 0.347
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 3.946780925991322e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}