| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.458889961242676, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5533, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7268170426065163, | |
| "eval_f1": 0.6326680574676724, | |
| "eval_loss": 0.5133728384971619, | |
| "eval_precision": 0.6605799373040753, | |
| "eval_recall": 0.6242044008001455, | |
| "eval_runtime": 1.774, | |
| "eval_samples_per_second": 224.915, | |
| "eval_steps_per_second": 28.185, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.8158535957336426, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4779, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7418546365914787, | |
| "eval_f1": 0.7122401394791937, | |
| "eval_loss": 0.4949621260166168, | |
| "eval_precision": 0.7053803339517626, | |
| "eval_recall": 0.734860883797054, | |
| "eval_runtime": 1.7746, | |
| "eval_samples_per_second": 224.844, | |
| "eval_steps_per_second": 28.176, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.156679630279541, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4097, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.7664715719063545, | |
| "eval_loss": 0.3772188425064087, | |
| "eval_precision": 0.8092877840475827, | |
| "eval_recall": 0.7458628841607565, | |
| "eval_runtime": 1.8118, | |
| "eval_samples_per_second": 220.228, | |
| "eval_steps_per_second": 27.597, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.239713191986084, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3451, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8169941409717701, | |
| "eval_loss": 0.3511227071285248, | |
| "eval_precision": 0.8104735988883742, | |
| "eval_recall": 0.8250591016548463, | |
| "eval_runtime": 1.8097, | |
| "eval_samples_per_second": 220.479, | |
| "eval_steps_per_second": 27.629, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.155226707458496, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2959, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8255172205802521, | |
| "eval_loss": 0.32013869285583496, | |
| "eval_precision": 0.8239495798319327, | |
| "eval_recall": 0.8271503909801782, | |
| "eval_runtime": 1.81, | |
| "eval_samples_per_second": 220.445, | |
| "eval_steps_per_second": 27.625, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.97943115234375, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2727, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8447157518450185, | |
| "eval_loss": 0.3176342844963074, | |
| "eval_precision": 0.8325401217487549, | |
| "eval_recall": 0.864248045099109, | |
| "eval_runtime": 1.8128, | |
| "eval_samples_per_second": 220.097, | |
| "eval_steps_per_second": 27.581, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.0954539775848389, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2595, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8524146298159436, | |
| "eval_loss": 0.2958522439002991, | |
| "eval_precision": 0.8451250578971746, | |
| "eval_recall": 0.8613384251682124, | |
| "eval_runtime": 1.8171, | |
| "eval_samples_per_second": 219.582, | |
| "eval_steps_per_second": 27.516, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 6.516312122344971, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2409, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8649122807017544, | |
| "eval_loss": 0.28329744935035706, | |
| "eval_precision": 0.8710116366366366, | |
| "eval_recall": 0.8594744498999818, | |
| "eval_runtime": 1.8171, | |
| "eval_samples_per_second": 219.577, | |
| "eval_steps_per_second": 27.516, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.709987163543701, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2298, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.850729517396184, | |
| "eval_loss": 0.2893889546394348, | |
| "eval_precision": 0.8535087719298247, | |
| "eval_recall": 0.8481087470449173, | |
| "eval_runtime": 1.8128, | |
| "eval_samples_per_second": 220.099, | |
| "eval_steps_per_second": 27.581, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.345912933349609, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2221, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8614765038536611, | |
| "eval_loss": 0.2884393632411957, | |
| "eval_precision": 0.8686536646744258, | |
| "eval_recall": 0.8552009456264775, | |
| "eval_runtime": 1.8154, | |
| "eval_samples_per_second": 219.786, | |
| "eval_steps_per_second": 27.542, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.3872387409210205, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1986, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8587719298245614, | |
| "eval_loss": 0.2855367362499237, | |
| "eval_precision": 0.864771021021021, | |
| "eval_recall": 0.8534278959810875, | |
| "eval_runtime": 1.8165, | |
| "eval_samples_per_second": 219.656, | |
| "eval_steps_per_second": 27.526, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 5.007177829742432, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1964, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8521068445832446, | |
| "eval_loss": 0.29210031032562256, | |
| "eval_precision": 0.8693800752624282, | |
| "eval_recall": 0.8391525731951264, | |
| "eval_runtime": 1.8152, | |
| "eval_samples_per_second": 219.815, | |
| "eval_steps_per_second": 27.546, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.4582099914550781, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1783, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8649122807017544, | |
| "eval_loss": 0.3103856146335602, | |
| "eval_precision": 0.8710116366366366, | |
| "eval_recall": 0.8594744498999818, | |
| "eval_runtime": 1.8138, | |
| "eval_samples_per_second": 219.982, | |
| "eval_steps_per_second": 27.567, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 6.738508701324463, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1788, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8689068100358424, | |
| "eval_loss": 0.3015482425689697, | |
| "eval_precision": 0.863953693884765, | |
| "eval_recall": 0.8744771776686671, | |
| "eval_runtime": 1.8152, | |
| "eval_samples_per_second": 219.806, | |
| "eval_steps_per_second": 27.545, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.6257522106170654, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.172, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8595070422535211, | |
| "eval_loss": 0.3011764883995056, | |
| "eval_precision": 0.8633733523114054, | |
| "eval_recall": 0.8559283506092017, | |
| "eval_runtime": 1.8131, | |
| "eval_samples_per_second": 220.062, | |
| "eval_steps_per_second": 27.577, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 4.563363552093506, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1563, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8695225637671682, | |
| "eval_loss": 0.3159307837486267, | |
| "eval_precision": 0.8631532846715328, | |
| "eval_recall": 0.8769776322967813, | |
| "eval_runtime": 1.8172, | |
| "eval_samples_per_second": 219.574, | |
| "eval_steps_per_second": 27.516, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.206107258796692, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1512, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8572517421602788, | |
| "eval_loss": 0.32489535212516785, | |
| "eval_precision": 0.8679426449878376, | |
| "eval_recall": 0.8484269867248591, | |
| "eval_runtime": 1.8131, | |
| "eval_samples_per_second": 220.063, | |
| "eval_steps_per_second": 27.577, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.8656744956970215, | |
| "learning_rate": 5e-06, | |
| "loss": 0.151, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.855319904024935, | |
| "eval_loss": 0.3245084285736084, | |
| "eval_precision": 0.862378106322743, | |
| "eval_recall": 0.8491543917075832, | |
| "eval_runtime": 1.8159, | |
| "eval_samples_per_second": 219.726, | |
| "eval_steps_per_second": 27.535, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.600020170211792, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1461, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8614765038536611, | |
| "eval_loss": 0.328171044588089, | |
| "eval_precision": 0.8686536646744258, | |
| "eval_recall": 0.8552009456264775, | |
| "eval_runtime": 1.8138, | |
| "eval_samples_per_second": 219.976, | |
| "eval_steps_per_second": 27.566, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.394913673400879, | |
| "learning_rate": 0.0, | |
| "loss": 0.1555, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8622036668943447, | |
| "eval_loss": 0.32483023405075073, | |
| "eval_precision": 0.8671602787456446, | |
| "eval_recall": 0.8577014002545917, | |
| "eval_runtime": 1.8125, | |
| "eval_samples_per_second": 220.138, | |
| "eval_steps_per_second": 27.586, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.2495564101172275, | |
| "train_runtime": 624.4298, | |
| "train_samples_per_second": 116.522, | |
| "train_steps_per_second": 3.908 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8444128359504000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |