trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 8304,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18063583815028902,
      "grad_norm": 7.173945903778076,
      "learning_rate": 9.397880539499037e-06,
      "loss": 6.2596,
      "step": 500
    },
    {
      "epoch": 0.36127167630057805,
      "grad_norm": 7.090274333953857,
      "learning_rate": 8.795761078998074e-06,
      "loss": 6.1274,
      "step": 1000
    },
    {
      "epoch": 0.541907514450867,
      "grad_norm": 6.882017612457275,
      "learning_rate": 8.19364161849711e-06,
      "loss": 6.0157,
      "step": 1500
    },
    {
      "epoch": 0.7225433526011561,
      "grad_norm": 6.622304916381836,
      "learning_rate": 7.591522157996147e-06,
      "loss": 5.9661,
      "step": 2000
    },
    {
      "epoch": 0.903179190751445,
      "grad_norm": 6.847517967224121,
      "learning_rate": 6.989402697495183e-06,
      "loss": 5.9061,
      "step": 2500
    },
    {
      "epoch": 1.083815028901734,
      "grad_norm": 6.756404399871826,
      "learning_rate": 6.387283236994221e-06,
      "loss": 5.8773,
      "step": 3000
    },
    {
      "epoch": 1.2644508670520231,
      "grad_norm": 6.702535629272461,
      "learning_rate": 5.785163776493257e-06,
      "loss": 5.8402,
      "step": 3500
    },
    {
      "epoch": 1.4450867052023122,
      "grad_norm": 6.539181709289551,
      "learning_rate": 5.1830443159922924e-06,
      "loss": 5.8124,
      "step": 4000
    },
    {
      "epoch": 1.6257225433526012,
      "grad_norm": 6.716819763183594,
      "learning_rate": 4.58092485549133e-06,
      "loss": 5.7907,
      "step": 4500
    },
    {
      "epoch": 1.80635838150289,
      "grad_norm": 6.560629367828369,
      "learning_rate": 3.978805394990367e-06,
      "loss": 5.769,
      "step": 5000
    },
    {
      "epoch": 1.9869942196531793,
      "grad_norm": 6.585288047790527,
      "learning_rate": 3.376685934489403e-06,
      "loss": 5.746,
      "step": 5500
    },
    {
      "epoch": 2.167630057803468,
      "grad_norm": 6.649946689605713,
      "learning_rate": 2.774566473988439e-06,
      "loss": 5.7399,
      "step": 6000
    },
    {
      "epoch": 2.348265895953757,
      "grad_norm": 6.707363605499268,
      "learning_rate": 2.1724470134874763e-06,
      "loss": 5.7306,
      "step": 6500
    },
    {
      "epoch": 2.5289017341040463,
      "grad_norm": 6.509890556335449,
      "learning_rate": 1.5703275529865126e-06,
      "loss": 5.7217,
      "step": 7000
    },
    {
      "epoch": 2.709537572254335,
      "grad_norm": 6.645078659057617,
      "learning_rate": 9.682080924855492e-07,
      "loss": 5.7249,
      "step": 7500
    },
    {
      "epoch": 2.8901734104046244,
      "grad_norm": 6.534315586090088,
      "learning_rate": 3.660886319845858e-07,
      "loss": 5.7126,
      "step": 8000
    },
    {
      "epoch": 3.0,
      "step": 8304,
      "total_flos": 1.7254148744798807e+18,
      "train_loss": 5.854142925872509,
      "train_runtime": 17646.7524,
      "train_samples_per_second": 15.055,
      "train_steps_per_second": 0.471
    }
  ],
  "logging_steps": 500,
  "max_steps": 8304,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7254148744798807e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}