| { | |
| "best_metric": 0.0016791470116004348, | |
| "best_model_checkpoint": "poisoned-spam-retest/checkpoint-204", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 204, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04411764705882353, | |
| "grad_norm": 30.36427116394043, | |
| "learning_rate": 2.3809523809523808e-06, | |
| "loss": 7.0693, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08823529411764706, | |
| "grad_norm": 35.84011459350586, | |
| "learning_rate": 9.523809523809523e-06, | |
| "loss": 6.9372, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1323529411764706, | |
| "grad_norm": 64.64055633544922, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 5.7856, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.17647058823529413, | |
| "grad_norm": 29.567256927490234, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 1.9984, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.22058823529411764, | |
| "grad_norm": 1.483197808265686, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.1307, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2647058823529412, | |
| "grad_norm": 0.03688773885369301, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.0038, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3088235294117647, | |
| "grad_norm": 0.010704166255891323, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 0.0005, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 0.009416253305971622, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0003, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.39705882352941174, | |
| "grad_norm": 4.293992519378662, | |
| "learning_rate": 4.918032786885246e-05, | |
| "loss": 0.3668, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4411764705882353, | |
| "grad_norm": 0.021360499784350395, | |
| "learning_rate": 4.836065573770492e-05, | |
| "loss": 0.0005, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4852941176470588, | |
| "grad_norm": 0.03743103891611099, | |
| "learning_rate": 4.754098360655738e-05, | |
| "loss": 0.0012, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.5294117647058824, | |
| "grad_norm": 0.060405198484659195, | |
| "learning_rate": 4.672131147540984e-05, | |
| "loss": 0.0017, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5735294117647058, | |
| "grad_norm": 0.03662671148777008, | |
| "learning_rate": 4.59016393442623e-05, | |
| "loss": 0.0012, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.6176470588235294, | |
| "grad_norm": 0.02251148410141468, | |
| "learning_rate": 4.508196721311476e-05, | |
| "loss": 0.0009, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6617647058823529, | |
| "grad_norm": 0.014444094151258469, | |
| "learning_rate": 4.426229508196721e-05, | |
| "loss": 0.0006, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 0.013513605110347271, | |
| "learning_rate": 4.3442622950819674e-05, | |
| "loss": 0.0005, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.01167203951627016, | |
| "learning_rate": 4.262295081967213e-05, | |
| "loss": 0.0004, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.7941176470588235, | |
| "grad_norm": 0.009744256734848022, | |
| "learning_rate": 4.1803278688524595e-05, | |
| "loss": 0.0004, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8382352941176471, | |
| "grad_norm": 0.007370511535555124, | |
| "learning_rate": 4.098360655737705e-05, | |
| "loss": 0.0003, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": 0.009272449649870396, | |
| "learning_rate": 4.016393442622951e-05, | |
| "loss": 0.3038, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9264705882352942, | |
| "grad_norm": 0.011772528290748596, | |
| "learning_rate": 3.934426229508197e-05, | |
| "loss": 0.0003, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.9705882352941176, | |
| "grad_norm": 0.017451414838433266, | |
| "learning_rate": 3.8524590163934424e-05, | |
| "loss": 0.0006, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9926470588235294, | |
| "eval_auc": 1.0, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.051786236464977264, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.1174, | |
| "eval_samples_per_second": 1158.243, | |
| "eval_steps_per_second": 76.648, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.0147058823529411, | |
| "grad_norm": 0.020196711644530296, | |
| "learning_rate": 3.7704918032786885e-05, | |
| "loss": 0.0007, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 0.022609353065490723, | |
| "learning_rate": 3.6885245901639346e-05, | |
| "loss": 0.0007, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.1029411764705883, | |
| "grad_norm": 0.023020507767796516, | |
| "learning_rate": 3.6065573770491806e-05, | |
| "loss": 0.0007, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.1470588235294117, | |
| "grad_norm": 0.016774509102106094, | |
| "learning_rate": 3.524590163934427e-05, | |
| "loss": 0.0006, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.1911764705882353, | |
| "grad_norm": 0.01388185378164053, | |
| "learning_rate": 3.442622950819672e-05, | |
| "loss": 0.0006, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.2352941176470589, | |
| "grad_norm": 0.01662955991923809, | |
| "learning_rate": 3.360655737704918e-05, | |
| "loss": 0.0006, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.2794117647058822, | |
| "grad_norm": 0.010529501363635063, | |
| "learning_rate": 3.2786885245901635e-05, | |
| "loss": 0.0004, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "grad_norm": 0.00965579692274332, | |
| "learning_rate": 3.19672131147541e-05, | |
| "loss": 0.0004, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.3676470588235294, | |
| "grad_norm": 0.009518979117274284, | |
| "learning_rate": 3.114754098360656e-05, | |
| "loss": 0.0003, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 0.008063753135502338, | |
| "learning_rate": 3.0327868852459017e-05, | |
| "loss": 0.0003, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.4558823529411764, | |
| "grad_norm": 0.0104528134688735, | |
| "learning_rate": 2.9508196721311478e-05, | |
| "loss": 0.0003, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.006017519161105156, | |
| "learning_rate": 2.8688524590163935e-05, | |
| "loss": 0.0002, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.5441176470588234, | |
| "grad_norm": 0.006928455550223589, | |
| "learning_rate": 2.7868852459016392e-05, | |
| "loss": 0.0003, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.5882352941176472, | |
| "grad_norm": 0.006857709027826786, | |
| "learning_rate": 2.7049180327868856e-05, | |
| "loss": 0.0002, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.6323529411764706, | |
| "grad_norm": 0.00756242498755455, | |
| "learning_rate": 2.6229508196721314e-05, | |
| "loss": 0.2468, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 1.6764705882352942, | |
| "grad_norm": 0.008452671580016613, | |
| "learning_rate": 2.540983606557377e-05, | |
| "loss": 0.0003, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.7205882352941178, | |
| "grad_norm": 0.013919142074882984, | |
| "learning_rate": 2.459016393442623e-05, | |
| "loss": 0.0004, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 0.011829485185444355, | |
| "learning_rate": 2.377049180327869e-05, | |
| "loss": 0.0004, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.8088235294117647, | |
| "grad_norm": 0.017397664487361908, | |
| "learning_rate": 2.295081967213115e-05, | |
| "loss": 0.0005, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 1.8529411764705883, | |
| "grad_norm": 0.03783185034990311, | |
| "learning_rate": 2.2131147540983607e-05, | |
| "loss": 0.1784, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.8970588235294117, | |
| "grad_norm": 0.10314859449863434, | |
| "learning_rate": 2.1311475409836064e-05, | |
| "loss": 0.0014, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 1.9411764705882353, | |
| "grad_norm": 0.027918625622987747, | |
| "learning_rate": 2.0491803278688525e-05, | |
| "loss": 0.0012, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.9852941176470589, | |
| "grad_norm": 0.011717199347913265, | |
| "learning_rate": 1.9672131147540985e-05, | |
| "loss": 0.0006, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9926470588235294, | |
| "eval_auc": 1.0, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.01687568612396717, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.1153, | |
| "eval_samples_per_second": 1179.688, | |
| "eval_steps_per_second": 78.068, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.0294117647058822, | |
| "grad_norm": 0.013379819691181183, | |
| "learning_rate": 1.8852459016393442e-05, | |
| "loss": 0.0005, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 2.073529411764706, | |
| "grad_norm": 0.01104449387639761, | |
| "learning_rate": 1.8032786885245903e-05, | |
| "loss": 0.0004, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.01606309972703457, | |
| "learning_rate": 1.721311475409836e-05, | |
| "loss": 0.0004, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.161764705882353, | |
| "grad_norm": 0.014542381279170513, | |
| "learning_rate": 1.6393442622950818e-05, | |
| "loss": 0.0005, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "grad_norm": 0.014963783323764801, | |
| "learning_rate": 1.557377049180328e-05, | |
| "loss": 0.0004, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.008563211187720299, | |
| "learning_rate": 1.4754098360655739e-05, | |
| "loss": 0.0003, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.2941176470588234, | |
| "grad_norm": 0.009410593658685684, | |
| "learning_rate": 1.3934426229508196e-05, | |
| "loss": 0.0003, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.338235294117647, | |
| "grad_norm": 0.009819586761295795, | |
| "learning_rate": 1.3114754098360657e-05, | |
| "loss": 0.0003, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 2.3823529411764706, | |
| "grad_norm": 1.7785907983779907, | |
| "learning_rate": 1.2295081967213116e-05, | |
| "loss": 0.0683, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 2.426470588235294, | |
| "grad_norm": 0.011425049044191837, | |
| "learning_rate": 1.1475409836065575e-05, | |
| "loss": 0.0003, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 0.00907618086785078, | |
| "learning_rate": 1.0655737704918032e-05, | |
| "loss": 0.0003, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 2.514705882352941, | |
| "grad_norm": 0.010396580211818218, | |
| "learning_rate": 9.836065573770493e-06, | |
| "loss": 0.0003, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 2.5588235294117645, | |
| "grad_norm": 0.012363788671791553, | |
| "learning_rate": 9.016393442622952e-06, | |
| "loss": 0.0003, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 2.6029411764705883, | |
| "grad_norm": 0.012369702570140362, | |
| "learning_rate": 8.196721311475409e-06, | |
| "loss": 0.0004, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 2.6470588235294117, | |
| "grad_norm": 0.011759784072637558, | |
| "learning_rate": 7.3770491803278695e-06, | |
| "loss": 0.0004, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.6911764705882355, | |
| "grad_norm": 0.010871393606066704, | |
| "learning_rate": 6.557377049180328e-06, | |
| "loss": 0.0003, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 2.735294117647059, | |
| "grad_norm": 0.011349204927682877, | |
| "learning_rate": 5.737704918032787e-06, | |
| "loss": 0.0003, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.7794117647058822, | |
| "grad_norm": 0.009962545707821846, | |
| "learning_rate": 4.918032786885246e-06, | |
| "loss": 0.0003, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.007712112739682198, | |
| "learning_rate": 4.098360655737704e-06, | |
| "loss": 0.0003, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.8676470588235294, | |
| "grad_norm": 0.007966699078679085, | |
| "learning_rate": 3.278688524590164e-06, | |
| "loss": 0.0003, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.911764705882353, | |
| "grad_norm": 0.00961599312722683, | |
| "learning_rate": 2.459016393442623e-06, | |
| "loss": 0.0003, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 2.9558823529411766, | |
| "grad_norm": 0.00812006276100874, | |
| "learning_rate": 1.639344262295082e-06, | |
| "loss": 0.0003, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.008510351181030273, | |
| "learning_rate": 8.19672131147541e-07, | |
| "loss": 0.0003, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 1.0, | |
| "eval_auc": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 0.0016791470116004348, | |
| "eval_precision": 1.0, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 0.113, | |
| "eval_samples_per_second": 1203.487, | |
| "eval_steps_per_second": 79.643, | |
| "step": 204 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 204, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.01 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 54046698651648.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |