trainer_state.json

{
  "best_metric": 0.48500654101371765,
  "best_model_checkpoint": "./xlam_1epoch_pythonic5e-6/checkpoint-56",
  "epoch": 0.9876543209876543,
  "eval_steps": 7,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11522633744855967,
      "eval_loss": 0.4905904531478882,
      "eval_runtime": 40.3827,
      "eval_samples_per_second": 1.882,
      "eval_steps_per_second": 1.882,
      "step": 7
    },
    {
      "epoch": 0.1646090534979424,
      "grad_norm": 6.2620391845703125,
      "learning_rate": 4.7689385491773934e-06,
      "loss": 8.4783,
      "step": 10
    },
    {
      "epoch": 0.23045267489711935,
      "eval_loss": 0.4900592863559723,
      "eval_runtime": 40.7901,
      "eval_samples_per_second": 1.863,
      "eval_steps_per_second": 1.863,
      "step": 14
    },
    {
      "epoch": 0.3292181069958848,
      "grad_norm": 6.400896072387695,
      "learning_rate": 3.9029676634059565e-06,
      "loss": 8.8473,
      "step": 20
    },
    {
      "epoch": 0.345679012345679,
      "eval_loss": 0.4889541566371918,
      "eval_runtime": 40.6095,
      "eval_samples_per_second": 1.871,
      "eval_steps_per_second": 1.871,
      "step": 21
    },
    {
      "epoch": 0.4609053497942387,
      "eval_loss": 0.4875749349594116,
      "eval_runtime": 40.776,
      "eval_samples_per_second": 1.864,
      "eval_steps_per_second": 1.864,
      "step": 28
    },
    {
      "epoch": 0.49382716049382713,
      "grad_norm": 7.524162769317627,
      "learning_rate": 2.6353472714635443e-06,
      "loss": 8.0414,
      "step": 30
    },
    {
      "epoch": 0.5761316872427984,
      "eval_loss": 0.48689982295036316,
      "eval_runtime": 40.5867,
      "eval_samples_per_second": 1.873,
      "eval_steps_per_second": 1.873,
      "step": 35
    },
    {
      "epoch": 0.6584362139917695,
      "grad_norm": 7.981804370880127,
      "learning_rate": 1.328978898250525e-06,
      "loss": 7.9097,
      "step": 40
    },
    {
      "epoch": 0.691358024691358,
      "eval_loss": 0.4860202968120575,
      "eval_runtime": 40.5514,
      "eval_samples_per_second": 1.874,
      "eval_steps_per_second": 1.874,
      "step": 42
    },
    {
      "epoch": 0.8065843621399177,
      "eval_loss": 0.4851381480693817,
      "eval_runtime": 40.6536,
      "eval_samples_per_second": 1.869,
      "eval_steps_per_second": 1.869,
      "step": 49
    },
    {
      "epoch": 0.823045267489712,
      "grad_norm": 6.65182638168335,
      "learning_rate": 3.578570595810274e-07,
      "loss": 7.7658,
      "step": 50
    },
    {
      "epoch": 0.9218106995884774,
      "eval_loss": 0.48500654101371765,
      "eval_runtime": 41.0413,
      "eval_samples_per_second": 1.852,
      "eval_steps_per_second": 1.852,
      "step": 56
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 8.849071502685547,
      "learning_rate": 0.0,
      "loss": 7.9953,
      "step": 60
    },
    {
      "epoch": 0.9876543209876543,
      "step": 60,
      "total_flos": 1.1414226576257188e+18,
      "train_loss": 8.17294947306315,
      "train_runtime": 9791.4602,
      "train_samples_per_second": 0.199,
      "train_steps_per_second": 0.006
    }
  ],
  "logging_steps": 10,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 7,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1414226576257188e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}