| { | |
| "best_metric": 0.7722007722007722, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-hq22ls8k/checkpoint-400", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 400, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.1560730934143066, | |
| "learning_rate": 9.426932053278163e-05, | |
| "loss": 0.595, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 5.519632339477539, | |
| "learning_rate": 9.267153543900568e-05, | |
| "loss": 0.5409, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 4.012811183929443, | |
| "learning_rate": 9.107375034522972e-05, | |
| "loss": 0.4944, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 9.410440444946289, | |
| "learning_rate": 8.947596525145376e-05, | |
| "loss": 0.416, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 2.913337230682373, | |
| "learning_rate": 8.78781801576778e-05, | |
| "loss": 0.4236, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 1.957322359085083, | |
| "learning_rate": 8.628039506390185e-05, | |
| "loss": 0.4094, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.4026949405670166, | |
| "learning_rate": 8.468260997012588e-05, | |
| "loss": 0.331, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.6281132698059082, | |
| "learning_rate": 8.308482487634992e-05, | |
| "loss": 0.3211, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 2.870759963989258, | |
| "learning_rate": 8.148703978257395e-05, | |
| "loss": 0.2933, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.9348390102386475, | |
| "learning_rate": 7.9889254688798e-05, | |
| "loss": 0.4921, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8375, | |
| "eval_f1": 0.7670250896057348, | |
| "eval_loss": 0.358794629573822, | |
| "eval_precision": 0.6858974358974359, | |
| "eval_recall": 0.8699186991869918, | |
| "eval_runtime": 1.5446, | |
| "eval_samples_per_second": 258.973, | |
| "eval_steps_per_second": 16.186, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 2.657993793487549, | |
| "learning_rate": 7.829146959502203e-05, | |
| "loss": 0.3661, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 9.199877738952637, | |
| "learning_rate": 7.669368450124608e-05, | |
| "loss": 0.3209, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 6.665186882019043, | |
| "learning_rate": 7.509589940747012e-05, | |
| "loss": 0.3953, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 5.048425197601318, | |
| "learning_rate": 7.349811431369416e-05, | |
| "loss": 0.2938, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 3.8985488414764404, | |
| "learning_rate": 7.19003292199182e-05, | |
| "loss": 0.2387, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.6217758655548096, | |
| "learning_rate": 7.030254412614223e-05, | |
| "loss": 0.2279, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 13.243719100952148, | |
| "learning_rate": 6.870475903236628e-05, | |
| "loss": 0.243, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 1.4423586130142212, | |
| "learning_rate": 6.710697393859032e-05, | |
| "loss": 0.3166, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 3.1215100288391113, | |
| "learning_rate": 6.550918884481436e-05, | |
| "loss": 0.3601, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.1280789375305176, | |
| "learning_rate": 6.39114037510384e-05, | |
| "loss": 0.2303, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.87, | |
| "eval_f1": 0.7699115044247787, | |
| "eval_loss": 0.30909162759780884, | |
| "eval_precision": 0.8446601941747572, | |
| "eval_recall": 0.7073170731707317, | |
| "eval_runtime": 1.5339, | |
| "eval_samples_per_second": 260.775, | |
| "eval_steps_per_second": 16.298, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 2.1738266944885254, | |
| "learning_rate": 6.231361865726245e-05, | |
| "loss": 0.1502, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 25.343351364135742, | |
| "learning_rate": 6.0715833563486475e-05, | |
| "loss": 0.1578, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 3.0342938899993896, | |
| "learning_rate": 5.9118048469710523e-05, | |
| "loss": 0.1892, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 15.658936500549316, | |
| "learning_rate": 5.752026337593456e-05, | |
| "loss": 0.2544, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 15.843140602111816, | |
| "learning_rate": 5.5922478282158606e-05, | |
| "loss": 0.2167, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 5.072413444519043, | |
| "learning_rate": 5.432469318838264e-05, | |
| "loss": 0.2226, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 5.797332763671875, | |
| "learning_rate": 5.272690809460668e-05, | |
| "loss": 0.1347, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 9.168190956115723, | |
| "learning_rate": 5.112912300083072e-05, | |
| "loss": 0.1976, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 5.497272968292236, | |
| "learning_rate": 4.9531337907054765e-05, | |
| "loss": 0.2177, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.3119521141052246, | |
| "learning_rate": 4.79335528132788e-05, | |
| "loss": 0.1684, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.845, | |
| "eval_f1": 0.743801652892562, | |
| "eval_loss": 0.47961556911468506, | |
| "eval_precision": 0.7563025210084033, | |
| "eval_recall": 0.7317073170731707, | |
| "eval_runtime": 1.5168, | |
| "eval_samples_per_second": 263.704, | |
| "eval_steps_per_second": 16.482, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 1.9762777090072632, | |
| "learning_rate": 4.633576771950284e-05, | |
| "loss": 0.1612, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 1.3597666025161743, | |
| "learning_rate": 4.473798262572688e-05, | |
| "loss": 0.0888, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 4.3360419273376465, | |
| "learning_rate": 4.3140197531950924e-05, | |
| "loss": 0.1549, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.6792251467704773, | |
| "learning_rate": 4.154241243817496e-05, | |
| "loss": 0.0805, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 4.299612998962402, | |
| "learning_rate": 3.9944627344399e-05, | |
| "loss": 0.1412, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 1.4792366027832031, | |
| "learning_rate": 3.834684225062304e-05, | |
| "loss": 0.0666, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 0.11450085788965225, | |
| "learning_rate": 3.674905715684708e-05, | |
| "loss": 0.1294, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.6610074639320374, | |
| "learning_rate": 3.515127206307112e-05, | |
| "loss": 0.0168, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 0.08138225972652435, | |
| "learning_rate": 3.355348696929516e-05, | |
| "loss": 0.1562, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 8.49350643157959, | |
| "learning_rate": 3.19557018755192e-05, | |
| "loss": 0.0895, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8525, | |
| "eval_f1": 0.7722007722007722, | |
| "eval_loss": 0.6465945243835449, | |
| "eval_precision": 0.7352941176470589, | |
| "eval_recall": 0.8130081300813008, | |
| "eval_runtime": 1.5173, | |
| "eval_samples_per_second": 263.626, | |
| "eval_steps_per_second": 16.477, | |
| "step": 400 | |
| } | |
| ], | |
| "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": 847261481803776.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 9.58671056265576e-05, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 6, | |
| "per_device_train_batch_size": 16, | |
| "seed": 14 | |
| } | |
| } | |