{
  "best_metric": 0.5153154134750366,
  "best_model_checkpoint": "/ML-A100/team/mm/eamon/self_instruction/seed_ppl/qwen14B_models/qwen_14B_d2_iter3_model/checkpoint-200",
  "epoch": 2.0,
  "eval_steps": 100,
  "global_step": 338,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05917159763313609,
      "grad_norm": 0.9615801643973331,
      "learning_rate": 2.5e-05,
      "loss": 1.8771,
      "step": 10
    },
    {
      "epoch": 0.11834319526627218,
      "grad_norm": 3.7252569579401107,
      "learning_rate": 5e-05,
      "loss": 1.4266,
      "step": 20
    },
    {
      "epoch": 0.17751479289940827,
      "grad_norm": 0.4587282095179799,
      "learning_rate": 4.98781004037916e-05,
      "loss": 0.7228,
      "step": 30
    },
    {
      "epoch": 0.23668639053254437,
      "grad_norm": 0.9607481645216482,
      "learning_rate": 4.951359037609088e-05,
      "loss": 0.5857,
      "step": 40
    },
    {
      "epoch": 0.2958579881656805,
      "grad_norm": 1.277269346490998,
      "learning_rate": 4.891002460691306e-05,
      "loss": 0.558,
      "step": 50
    },
    {
      "epoch": 0.35502958579881655,
      "grad_norm": 1.2676470043435204,
      "learning_rate": 4.807328905014201e-05,
      "loss": 0.6275,
      "step": 60
    },
    {
      "epoch": 0.41420118343195267,
      "grad_norm": 0.43321439772469345,
      "learning_rate": 4.7011543523897996e-05,
      "loss": 0.5004,
      "step": 70
    },
    {
      "epoch": 0.47337278106508873,
      "grad_norm": 0.8638153501814058,
      "learning_rate": 4.573514213625505e-05,
      "loss": 0.5087,
      "step": 80
    },
    {
      "epoch": 0.5325443786982249,
      "grad_norm": 3.2295994507917754,
      "learning_rate": 4.425653231231344e-05,
      "loss": 0.5459,
      "step": 90
    },
    {
      "epoch": 0.591715976331361,
      "grad_norm": 1.0832832075215777,
      "learning_rate": 4.259013340731224e-05,
      "loss": 0.4759,
      "step": 100
    },
    {
      "epoch": 0.591715976331361,
      "eval_loss": 0.5315915942192078,
      "eval_runtime": 20.3569,
      "eval_samples_per_second": 14.737,
      "eval_steps_per_second": 1.867,
      "step": 100
    },
    {
      "epoch": 0.650887573964497,
      "grad_norm": 1.1344737325339866,
      "learning_rate": 4.075219608954278e-05,
      "loss": 0.4617,
      "step": 110
    },
    {
      "epoch": 0.7100591715976331,
      "grad_norm": 1.0747858639409258,
      "learning_rate": 3.876064386435646e-05,
      "loss": 0.4369,
      "step": 120
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.4410328402897703,
      "learning_rate": 3.663489828471953e-05,
      "loss": 0.4933,
      "step": 130
    },
    {
      "epoch": 0.8284023668639053,
      "grad_norm": 1.306125146854077,
      "learning_rate": 3.4395689552855955e-05,
      "loss": 0.4431,
      "step": 140
    },
    {
      "epoch": 0.8875739644970414,
      "grad_norm": 1.3505699464811827,
      "learning_rate": 3.206485435998498e-05,
      "loss": 0.4063,
      "step": 150
    },
    {
      "epoch": 0.9467455621301775,
      "grad_norm": 1.158959539769833,
      "learning_rate": 2.9665122935613727e-05,
      "loss": 0.4936,
      "step": 160
    },
    {
      "epoch": 1.0059171597633136,
      "grad_norm": 0.686278368841292,
      "learning_rate": 2.7219897383073373e-05,
      "loss": 0.4633,
      "step": 170
    },
    {
      "epoch": 1.0650887573964498,
      "grad_norm": 0.7329855489762843,
      "learning_rate": 2.475302346296336e-05,
      "loss": 0.3889,
      "step": 180
    },
    {
      "epoch": 1.1242603550295858,
      "grad_norm": 1.511269029215083,
      "learning_rate": 2.2288558050064367e-05,
      "loss": 0.457,
      "step": 190
    },
    {
      "epoch": 1.183431952662722,
      "grad_norm": 0.7905550148998332,
      "learning_rate": 1.9850534531472546e-05,
      "loss": 0.3967,
      "step": 200
    },
    {
      "epoch": 1.183431952662722,
      "eval_loss": 0.5153154134750366,
      "eval_runtime": 18.9227,
      "eval_samples_per_second": 15.854,
      "eval_steps_per_second": 2.008,
      "step": 200
    },
    {
      "epoch": 1.242603550295858,
      "grad_norm": 1.5247295502601703,
      "learning_rate": 1.746272843378493e-05,
      "loss": 0.3733,
      "step": 210
    },
    {
      "epoch": 1.301775147928994,
      "grad_norm": 0.8404429082517917,
      "learning_rate": 1.5148425564932084e-05,
      "loss": 0.3933,
      "step": 220
    },
    {
      "epoch": 1.3609467455621302,
      "grad_norm": 1.064902485845328,
      "learning_rate": 1.2930194931731382e-05,
      "loss": 0.3892,
      "step": 230
    },
    {
      "epoch": 1.4201183431952662,
      "grad_norm": 0.8352671361637931,
      "learning_rate": 1.0829668647661559e-05,
      "loss": 0.3891,
      "step": 240
    },
    {
      "epoch": 1.4792899408284024,
      "grad_norm": 0.8368737776004409,
      "learning_rate": 8.867330977190877e-06,
      "loss": 0.3597,
      "step": 250
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.4958411343194272,
      "learning_rate": 7.062318573891716e-06,
      "loss": 0.4077,
      "step": 260
    },
    {
      "epoch": 1.5976331360946747,
      "grad_norm": 0.4641500517420258,
      "learning_rate": 5.4322338604131715e-06,
      "loss": 0.4247,
      "step": 270
    },
    {
      "epoch": 1.6568047337278107,
      "grad_norm": 0.8611324943813823,
      "learning_rate": 3.992973370223896e-06,
      "loss": 0.3805,
      "step": 280
    },
    {
      "epoch": 1.7159763313609466,
      "grad_norm": 1.2247521849952663,
      "learning_rate": 2.75857272513132e-06,
      "loss": 0.4078,
      "step": 290
    },
    {
      "epoch": 1.7751479289940828,
      "grad_norm": 1.2501388237565647,
      "learning_rate": 1.7410697603511383e-06,
      "loss": 0.3547,
      "step": 300
    },
    {
      "epoch": 1.7751479289940828,
      "eval_loss": 0.5177696347236633,
      "eval_runtime": 18.9794,
      "eval_samples_per_second": 15.807,
      "eval_steps_per_second": 2.002,
      "step": 300
    },
    {
      "epoch": 1.834319526627219,
      "grad_norm": 0.749968665840715,
      "learning_rate": 9.503871319271551e-07,
      "loss": 0.35,
      "step": 310
    },
    {
      "epoch": 1.893491124260355,
      "grad_norm": 0.9424148400717701,
      "learning_rate": 3.9423555131007925e-07,
      "loss": 0.3912,
      "step": 320
    },
    {
      "epoch": 1.952662721893491,
      "grad_norm": 0.8186461379045846,
      "learning_rate": 7.803859074854425e-08,
      "loss": 0.4186,
      "step": 330
    },
    {
      "epoch": 2.0,
      "step": 338,
      "total_flos": 84346414039040.0,
      "train_loss": 0.5209799306632499,
      "train_runtime": 1210.5031,
      "train_samples_per_second": 4.461,
      "train_steps_per_second": 0.279
    }
  ],
  "logging_steps": 10,
  "max_steps": 338,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 84346414039040.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}