last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004565167769915545,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00045651677699155445,
      "grad_norm": 2.7275002002716064,
      "learning_rate": 2e-05,
      "loss": 2.7206,
      "step": 1
    },
    {
      "epoch": 0.00045651677699155445,
      "eval_loss": 0.9178780913352966,
      "eval_runtime": 28.7385,
      "eval_samples_per_second": 32.117,
      "eval_steps_per_second": 16.076,
      "step": 1
    },
    {
      "epoch": 0.0009130335539831089,
      "grad_norm": 3.5982139110565186,
      "learning_rate": 4e-05,
      "loss": 3.7009,
      "step": 2
    },
    {
      "epoch": 0.0013695503309746632,
      "grad_norm": 3.0312438011169434,
      "learning_rate": 6e-05,
      "loss": 4.3561,
      "step": 3
    },
    {
      "epoch": 0.0013695503309746632,
      "eval_loss": 0.9134677648544312,
      "eval_runtime": 28.7171,
      "eval_samples_per_second": 32.141,
      "eval_steps_per_second": 16.088,
      "step": 3
    },
    {
      "epoch": 0.0018260671079662178,
      "grad_norm": 4.0302581787109375,
      "learning_rate": 8e-05,
      "loss": 4.7532,
      "step": 4
    },
    {
      "epoch": 0.0022825838849577723,
      "grad_norm": 3.8156661987304688,
      "learning_rate": 0.0001,
      "loss": 3.4387,
      "step": 5
    },
    {
      "epoch": 0.0027391006619493265,
      "grad_norm": 4.494126319885254,
      "learning_rate": 0.00012,
      "loss": 3.3432,
      "step": 6
    },
    {
      "epoch": 0.0027391006619493265,
      "eval_loss": 0.7736297845840454,
      "eval_runtime": 28.8536,
      "eval_samples_per_second": 31.989,
      "eval_steps_per_second": 16.012,
      "step": 6
    },
    {
      "epoch": 0.003195617438940881,
      "grad_norm": 3.811544418334961,
      "learning_rate": 0.00014,
      "loss": 2.5571,
      "step": 7
    },
    {
      "epoch": 0.0036521342159324356,
      "grad_norm": 5.701013565063477,
      "learning_rate": 0.00016,
      "loss": 2.3771,
      "step": 8
    },
    {
      "epoch": 0.00410865099292399,
      "grad_norm": 4.322719573974609,
      "learning_rate": 0.00018,
      "loss": 1.28,
      "step": 9
    },
    {
      "epoch": 0.00410865099292399,
      "eval_loss": 0.3839247226715088,
      "eval_runtime": 28.854,
      "eval_samples_per_second": 31.989,
      "eval_steps_per_second": 16.012,
      "step": 9
    },
    {
      "epoch": 0.004565167769915545,
      "grad_norm": 4.11519193649292,
      "learning_rate": 0.0002,
      "loss": 2.1208,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1440390402785280.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}