| { | |
| "best_metric": 0.782258064516129, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-uiccwvov/checkpoint-300", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.3594228029251099, | |
| "learning_rate": 8.203619307998052e-06, | |
| "loss": 0.6733, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.288736343383789, | |
| "learning_rate": 8.064574912947238e-06, | |
| "loss": 0.5727, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.9693996906280518, | |
| "learning_rate": 7.925530517896423e-06, | |
| "loss": 0.5586, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.1887316703796387, | |
| "learning_rate": 7.78648612284561e-06, | |
| "loss": 0.5175, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 2.667508602142334, | |
| "learning_rate": 7.647441727794794e-06, | |
| "loss": 0.4807, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 7.1765947341918945, | |
| "learning_rate": 7.50839733274398e-06, | |
| "loss": 0.5072, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 3.7157256603240967, | |
| "learning_rate": 7.369352937693165e-06, | |
| "loss": 0.48, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.5790538787841797, | |
| "learning_rate": 7.2303085426423516e-06, | |
| "loss": 0.4395, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 4.587155342102051, | |
| "learning_rate": 7.091264147591536e-06, | |
| "loss": 0.5404, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.335803508758545, | |
| "learning_rate": 6.952219752540723e-06, | |
| "loss": 0.4595, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7775, | |
| "eval_f1": 0.6666666666666666, | |
| "eval_loss": 0.4462689161300659, | |
| "eval_precision": 0.6180555555555556, | |
| "eval_recall": 0.7235772357723578, | |
| "eval_runtime": 1.5218, | |
| "eval_samples_per_second": 262.839, | |
| "eval_steps_per_second": 16.427, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 2.8303627967834473, | |
| "learning_rate": 6.8131753574899075e-06, | |
| "loss": 0.4362, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.731326103210449, | |
| "learning_rate": 6.674130962439094e-06, | |
| "loss": 0.3568, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 3.0789365768432617, | |
| "learning_rate": 6.535086567388279e-06, | |
| "loss": 0.3805, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 6.227985382080078, | |
| "learning_rate": 6.396042172337465e-06, | |
| "loss": 0.3814, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 19.28496551513672, | |
| "learning_rate": 6.25699777728665e-06, | |
| "loss": 0.3048, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 7.075613975524902, | |
| "learning_rate": 6.117953382235835e-06, | |
| "loss": 0.3661, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 14.289627075195312, | |
| "learning_rate": 5.978908987185022e-06, | |
| "loss": 0.3559, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 3.69010066986084, | |
| "learning_rate": 5.8398645921342065e-06, | |
| "loss": 0.3944, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 4.366544723510742, | |
| "learning_rate": 5.700820197083393e-06, | |
| "loss": 0.2829, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.2513275146484375, | |
| "learning_rate": 5.561775802032578e-06, | |
| "loss": 0.3514, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8275, | |
| "eval_f1": 0.7526881720430108, | |
| "eval_loss": 0.36655348539352417, | |
| "eval_precision": 0.6730769230769231, | |
| "eval_recall": 0.8536585365853658, | |
| "eval_runtime": 1.5163, | |
| "eval_samples_per_second": 263.798, | |
| "eval_steps_per_second": 16.487, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 9.200742721557617, | |
| "learning_rate": 5.422731406981764e-06, | |
| "loss": 0.3214, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 5.735116004943848, | |
| "learning_rate": 5.283687011930949e-06, | |
| "loss": 0.2704, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 1.3424041271209717, | |
| "learning_rate": 5.144642616880135e-06, | |
| "loss": 0.2391, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 8.352746963500977, | |
| "learning_rate": 5.00559822182932e-06, | |
| "loss": 0.2979, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.4383323192596436, | |
| "learning_rate": 4.866553826778506e-06, | |
| "loss": 0.307, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 11.764205932617188, | |
| "learning_rate": 4.727509431727691e-06, | |
| "loss": 0.2697, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 6.212594985961914, | |
| "learning_rate": 4.588465036676878e-06, | |
| "loss": 0.2888, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 1.2262153625488281, | |
| "learning_rate": 4.449420641626062e-06, | |
| "loss": 0.3025, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 4.729959964752197, | |
| "learning_rate": 4.310376246575249e-06, | |
| "loss": 0.2624, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.7858563661575317, | |
| "learning_rate": 4.1713318515244335e-06, | |
| "loss": 0.3098, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.865, | |
| "eval_f1": 0.782258064516129, | |
| "eval_loss": 0.335612952709198, | |
| "eval_precision": 0.776, | |
| "eval_recall": 0.7886178861788617, | |
| "eval_runtime": 1.5218, | |
| "eval_samples_per_second": 262.855, | |
| "eval_steps_per_second": 16.428, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 635446111352832.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 8.342663703048867e-06, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 6, | |
| "per_device_train_batch_size": 16, | |
| "seed": 18 | |
| } | |
| } | |