| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6711409395973155, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013422818791946308, | |
| "grad_norm": 0.6845026612281799, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2256, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.013422818791946308, | |
| "eval_loss": 1.3543461561203003, | |
| "eval_runtime": 3.1855, | |
| "eval_samples_per_second": 15.696, | |
| "eval_steps_per_second": 4.081, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.026845637583892617, | |
| "grad_norm": 0.4787723422050476, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2686, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.040268456375838924, | |
| "grad_norm": 0.9371090531349182, | |
| "learning_rate": 9.990365154573717e-05, | |
| "loss": 1.2807, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.053691275167785234, | |
| "grad_norm": 0.5610334277153015, | |
| "learning_rate": 9.961501876182148e-05, | |
| "loss": 1.1412, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06711409395973154, | |
| "grad_norm": 0.7126432061195374, | |
| "learning_rate": 9.913533761814537e-05, | |
| "loss": 1.0747, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.08053691275167785, | |
| "grad_norm": 0.44282087683677673, | |
| "learning_rate": 9.846666218300807e-05, | |
| "loss": 1.0108, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09395973154362416, | |
| "grad_norm": 0.4470914602279663, | |
| "learning_rate": 9.761185582727977e-05, | |
| "loss": 0.9648, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.10738255033557047, | |
| "grad_norm": 0.4001728892326355, | |
| "learning_rate": 9.657457896300791e-05, | |
| "loss": 0.9402, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.12080536912751678, | |
| "grad_norm": 0.2786104381084442, | |
| "learning_rate": 9.535927336897098e-05, | |
| "loss": 0.8941, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1342281879194631, | |
| "grad_norm": 0.3536582589149475, | |
| "learning_rate": 9.397114317029975e-05, | |
| "loss": 0.8764, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1476510067114094, | |
| "grad_norm": 0.24438385665416718, | |
| "learning_rate": 9.241613255361455e-05, | |
| "loss": 0.8537, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1610738255033557, | |
| "grad_norm": 0.2583913803100586, | |
| "learning_rate": 9.070090031310558e-05, | |
| "loss": 0.8486, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.174496644295302, | |
| "grad_norm": 0.2444184273481369, | |
| "learning_rate": 8.883279133655399e-05, | |
| "loss": 0.8429, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.18791946308724833, | |
| "grad_norm": 0.20378562808036804, | |
| "learning_rate": 8.681980515339464e-05, | |
| "loss": 0.8203, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.20134228187919462, | |
| "grad_norm": 0.19167418777942657, | |
| "learning_rate": 8.467056167950311e-05, | |
| "loss": 0.8013, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.21476510067114093, | |
| "grad_norm": 0.1807529330253601, | |
| "learning_rate": 8.239426430539243e-05, | |
| "loss": 0.8229, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.22818791946308725, | |
| "grad_norm": 0.14956992864608765, | |
| "learning_rate": 8.000066048588211e-05, | |
| "loss": 0.8126, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.24161073825503357, | |
| "grad_norm": 0.13678358495235443, | |
| "learning_rate": 7.75e-05, | |
| "loss": 0.8127, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2550335570469799, | |
| "grad_norm": 0.15046660602092743, | |
| "learning_rate": 7.490299105985507e-05, | |
| "loss": 0.7896, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2684563758389262, | |
| "grad_norm": 0.14206001162528992, | |
| "learning_rate": 7.222075445642904e-05, | |
| "loss": 0.7804, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.28187919463087246, | |
| "grad_norm": 0.1394774317741394, | |
| "learning_rate": 6.946477593864228e-05, | |
| "loss": 0.8066, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2953020134228188, | |
| "grad_norm": 0.1308431327342987, | |
| "learning_rate": 6.664685702961344e-05, | |
| "loss": 0.8039, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.3087248322147651, | |
| "grad_norm": 0.12986968457698822, | |
| "learning_rate": 6.377906449072578e-05, | |
| "loss": 0.7952, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3221476510067114, | |
| "grad_norm": 0.13289688527584076, | |
| "learning_rate": 6.087367864990233e-05, | |
| "loss": 0.7903, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.33557046979865773, | |
| "grad_norm": 0.1467195302248001, | |
| "learning_rate": 5.794314081535644e-05, | |
| "loss": 0.8507, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.33557046979865773, | |
| "eval_loss": 0.7909823060035706, | |
| "eval_runtime": 3.2001, | |
| "eval_samples_per_second": 15.624, | |
| "eval_steps_per_second": 4.062, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.348993288590604, | |
| "grad_norm": 0.16195999085903168, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.7858, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3624161073825503, | |
| "grad_norm": 0.1427437961101532, | |
| "learning_rate": 5.205685918464356e-05, | |
| "loss": 0.7842, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.37583892617449666, | |
| "grad_norm": 0.1225617453455925, | |
| "learning_rate": 4.912632135009769e-05, | |
| "loss": 0.7713, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.38926174496644295, | |
| "grad_norm": 0.1048019751906395, | |
| "learning_rate": 4.6220935509274235e-05, | |
| "loss": 0.7604, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.40268456375838924, | |
| "grad_norm": 0.10601115971803665, | |
| "learning_rate": 4.3353142970386564e-05, | |
| "loss": 0.7612, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4161073825503356, | |
| "grad_norm": 0.10506386309862137, | |
| "learning_rate": 4.053522406135775e-05, | |
| "loss": 0.7592, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.42953020134228187, | |
| "grad_norm": 0.11626201122999191, | |
| "learning_rate": 3.777924554357096e-05, | |
| "loss": 0.7672, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.4429530201342282, | |
| "grad_norm": 0.10783777385950089, | |
| "learning_rate": 3.509700894014496e-05, | |
| "loss": 0.755, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.4563758389261745, | |
| "grad_norm": 0.11628258228302002, | |
| "learning_rate": 3.250000000000001e-05, | |
| "loss": 0.7495, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.4697986577181208, | |
| "grad_norm": 0.10531030595302582, | |
| "learning_rate": 2.9999339514117912e-05, | |
| "loss": 0.7676, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.48322147651006714, | |
| "grad_norm": 0.10979268699884415, | |
| "learning_rate": 2.760573569460757e-05, | |
| "loss": 0.793, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4966442953020134, | |
| "grad_norm": 0.11333921551704407, | |
| "learning_rate": 2.53294383204969e-05, | |
| "loss": 0.7676, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.5100671140939598, | |
| "grad_norm": 0.12159935384988785, | |
| "learning_rate": 2.3180194846605367e-05, | |
| "loss": 0.7693, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.5234899328859061, | |
| "grad_norm": 0.11211013048887253, | |
| "learning_rate": 2.1167208663446025e-05, | |
| "loss": 0.7542, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5369127516778524, | |
| "grad_norm": 0.1045965701341629, | |
| "learning_rate": 1.9299099686894423e-05, | |
| "loss": 0.7526, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5503355704697986, | |
| "grad_norm": 0.10266979038715363, | |
| "learning_rate": 1.758386744638546e-05, | |
| "loss": 0.7718, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.5637583892617449, | |
| "grad_norm": 0.09563658386468887, | |
| "learning_rate": 1.602885682970026e-05, | |
| "loss": 0.722, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5771812080536913, | |
| "grad_norm": 0.09837043285369873, | |
| "learning_rate": 1.464072663102903e-05, | |
| "loss": 0.7262, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.5906040268456376, | |
| "grad_norm": 0.09940379858016968, | |
| "learning_rate": 1.3425421036992098e-05, | |
| "loss": 0.7409, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.6040268456375839, | |
| "grad_norm": 0.10937907546758652, | |
| "learning_rate": 1.2388144172720251e-05, | |
| "loss": 0.7488, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6174496644295302, | |
| "grad_norm": 0.102512888610363, | |
| "learning_rate": 1.1533337816991932e-05, | |
| "loss": 0.7414, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.6308724832214765, | |
| "grad_norm": 0.10500387847423553, | |
| "learning_rate": 1.0864662381854632e-05, | |
| "loss": 0.7486, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.6442953020134228, | |
| "grad_norm": 0.10852666199207306, | |
| "learning_rate": 1.0384981238178534e-05, | |
| "loss": 0.7587, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6577181208053692, | |
| "grad_norm": 0.11192145198583603, | |
| "learning_rate": 1.0096348454262845e-05, | |
| "loss": 0.7629, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.6711409395973155, | |
| "grad_norm": 0.12789976596832275, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7869, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6711409395973155, | |
| "eval_loss": 0.7637184858322144, | |
| "eval_runtime": 3.1762, | |
| "eval_samples_per_second": 15.742, | |
| "eval_steps_per_second": 4.093, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.31191586550186e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |