| { | |
| "best_metric": 0.5581375956535339, | |
| "best_model_checkpoint": "output_pipe/H3K4me3/origin/checkpoint-800", | |
| "epoch": 4.0, | |
| "eval_steps": 200, | |
| "global_step": 1840, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21739130434782608, | |
| "grad_norm": 2.1376545429229736, | |
| "learning_rate": 2.9162011173184356e-05, | |
| "loss": 0.6597, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 1.4405479431152344, | |
| "learning_rate": 2.7486033519553074e-05, | |
| "loss": 0.6117, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "eval_accuracy": 0.6747282608695652, | |
| "eval_f1": 0.6735077482392049, | |
| "eval_loss": 0.5982626676559448, | |
| "eval_matthews_correlation": 0.3470163967739955, | |
| "eval_precision": 0.6734908378464989, | |
| "eval_recall": 0.673525560664697, | |
| "eval_runtime": 2.232, | |
| "eval_samples_per_second": 1648.739, | |
| "eval_steps_per_second": 25.986, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 1.6685551404953003, | |
| "learning_rate": 2.5810055865921788e-05, | |
| "loss": 0.5935, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 1.990510106086731, | |
| "learning_rate": 2.4134078212290505e-05, | |
| "loss": 0.5809, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_accuracy": 0.6880434782608695, | |
| "eval_f1": 0.6871193269344709, | |
| "eval_loss": 0.6194872856140137, | |
| "eval_matthews_correlation": 0.3743185807423757, | |
| "eval_precision": 0.6870041004036465, | |
| "eval_recall": 0.6873146091269535, | |
| "eval_runtime": 2.2292, | |
| "eval_samples_per_second": 1650.784, | |
| "eval_steps_per_second": 26.018, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "grad_norm": 2.2024917602539062, | |
| "learning_rate": 2.245810055865922e-05, | |
| "loss": 0.5567, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.3043478260869565, | |
| "grad_norm": 2.437084197998047, | |
| "learning_rate": 2.0782122905027933e-05, | |
| "loss": 0.5195, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3043478260869565, | |
| "eval_accuracy": 0.7046195652173913, | |
| "eval_f1": 0.7046146575420862, | |
| "eval_loss": 0.579476535320282, | |
| "eval_matthews_correlation": 0.4154010589723203, | |
| "eval_precision": 0.7078128585120804, | |
| "eval_recall": 0.7075882611774167, | |
| "eval_runtime": 2.231, | |
| "eval_samples_per_second": 1649.496, | |
| "eval_steps_per_second": 25.997, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.5217391304347827, | |
| "grad_norm": 2.1971843242645264, | |
| "learning_rate": 1.9106145251396648e-05, | |
| "loss": 0.5142, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 2.499269485473633, | |
| "learning_rate": 1.7430167597765365e-05, | |
| "loss": 0.5117, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_accuracy": 0.7152173913043478, | |
| "eval_f1": 0.7152031750167114, | |
| "eval_loss": 0.5581375956535339, | |
| "eval_matthews_correlation": 0.43725397945520167, | |
| "eval_precision": 0.7188413547237077, | |
| "eval_recall": 0.7184128347116946, | |
| "eval_runtime": 2.2287, | |
| "eval_samples_per_second": 1651.206, | |
| "eval_steps_per_second": 26.024, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.9565217391304348, | |
| "grad_norm": 2.400865316390991, | |
| "learning_rate": 1.575418994413408e-05, | |
| "loss": 0.4958, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.1739130434782608, | |
| "grad_norm": 3.1249494552612305, | |
| "learning_rate": 1.4078212290502793e-05, | |
| "loss": 0.3741, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.1739130434782608, | |
| "eval_accuracy": 0.7089673913043478, | |
| "eval_f1": 0.7089347006202478, | |
| "eval_loss": 0.631187379360199, | |
| "eval_matthews_correlation": 0.4216199019944169, | |
| "eval_precision": 0.7105799599109461, | |
| "eval_recall": 0.7110401932749887, | |
| "eval_runtime": 2.2302, | |
| "eval_samples_per_second": 1650.048, | |
| "eval_steps_per_second": 26.006, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.391304347826087, | |
| "grad_norm": 3.837869882583618, | |
| "learning_rate": 1.2402234636871509e-05, | |
| "loss": 0.3192, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "grad_norm": 5.140067100524902, | |
| "learning_rate": 1.0726256983240223e-05, | |
| "loss": 0.3017, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "eval_accuracy": 0.7027173913043478, | |
| "eval_f1": 0.7022740913558971, | |
| "eval_loss": 0.6591957807540894, | |
| "eval_matthews_correlation": 0.405317483978006, | |
| "eval_precision": 0.7022928277342106, | |
| "eval_recall": 0.7030253181221395, | |
| "eval_runtime": 2.2281, | |
| "eval_samples_per_second": 1651.626, | |
| "eval_steps_per_second": 26.031, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.8260869565217392, | |
| "grad_norm": 4.443952560424805, | |
| "learning_rate": 9.050279329608939e-06, | |
| "loss": 0.3041, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.0434782608695654, | |
| "grad_norm": 3.8865654468536377, | |
| "learning_rate": 7.374301675977653e-06, | |
| "loss": 0.2614, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.0434782608695654, | |
| "eval_accuracy": 0.7038043478260869, | |
| "eval_f1": 0.7020044568579538, | |
| "eval_loss": 0.8777711391448975, | |
| "eval_matthews_correlation": 0.40436111238724043, | |
| "eval_precision": 0.7026796058183274, | |
| "eval_recall": 0.701682735358198, | |
| "eval_runtime": 2.2272, | |
| "eval_samples_per_second": 1652.28, | |
| "eval_steps_per_second": 26.041, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.260869565217391, | |
| "grad_norm": 4.834942817687988, | |
| "learning_rate": 5.698324022346369e-06, | |
| "loss": 0.0812, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 9.44428539276123, | |
| "learning_rate": 4.022346368715084e-06, | |
| "loss": 0.0619, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "eval_accuracy": 0.7035326086956522, | |
| "eval_f1": 0.7035166487208262, | |
| "eval_loss": 1.1049858331680298, | |
| "eval_matthews_correlation": 0.4112438158939813, | |
| "eval_precision": 0.7054566698766234, | |
| "eval_recall": 0.7057872789097089, | |
| "eval_runtime": 5.1617, | |
| "eval_samples_per_second": 712.944, | |
| "eval_steps_per_second": 11.237, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.6956521739130435, | |
| "grad_norm": 11.50954532623291, | |
| "learning_rate": 2.3798882681564246e-06, | |
| "loss": 0.0753, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.9130434782608696, | |
| "grad_norm": 14.563551902770996, | |
| "learning_rate": 7.039106145251397e-07, | |
| "loss": 0.0612, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.9130434782608696, | |
| "eval_accuracy": 0.7021739130434783, | |
| "eval_f1": 0.7019710956416465, | |
| "eval_loss": 1.108843445777893, | |
| "eval_matthews_correlation": 0.40575281311869543, | |
| "eval_precision": 0.702496340032541, | |
| "eval_recall": 0.70325718643517, | |
| "eval_runtime": 2.2289, | |
| "eval_samples_per_second": 1651.071, | |
| "eval_steps_per_second": 26.022, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1840, | |
| "total_flos": 2.998630390984884e+16, | |
| "train_loss": 0.3753926209781481, | |
| "train_runtime": 306.3507, | |
| "train_samples_per_second": 384.383, | |
| "train_steps_per_second": 6.006 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1840, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.998630390984884e+16, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |