| { | |
| "best_metric": 3.282808780670166, | |
| "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_recent/mlm_unmasking/fr_mlm_new/wikipedia_30/checkpoint-44000", | |
| "epoch": 51.43191116306254, | |
| "eval_steps": 2000, | |
| "global_step": 44000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.33781414377557, | |
| "eval_loss": 6.900431156158447, | |
| "eval_runtime": 1.1096, | |
| "eval_samples_per_second": 1285.131, | |
| "eval_steps_per_second": 81.109, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.67562828755114, | |
| "grad_norm": 0.7734344005584717, | |
| "learning_rate": 4e-05, | |
| "loss": 7.3814, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.67562828755114, | |
| "eval_loss": 6.925902366638184, | |
| "eval_runtime": 1.0887, | |
| "eval_samples_per_second": 1309.827, | |
| "eval_steps_per_second": 82.668, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.01344243132671, | |
| "eval_loss": 6.909808158874512, | |
| "eval_runtime": 1.0952, | |
| "eval_samples_per_second": 1302.036, | |
| "eval_steps_per_second": 82.176, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.35125657510228, | |
| "grad_norm": 1.7065298557281494, | |
| "learning_rate": 8e-05, | |
| "loss": 6.845, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.35125657510228, | |
| "eval_loss": 6.668918132781982, | |
| "eval_runtime": 1.0886, | |
| "eval_samples_per_second": 1309.937, | |
| "eval_steps_per_second": 82.675, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 11.689070718877849, | |
| "eval_loss": 5.787526607513428, | |
| "eval_runtime": 1.0937, | |
| "eval_samples_per_second": 1303.854, | |
| "eval_steps_per_second": 82.291, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 14.02688486265342, | |
| "grad_norm": 3.4086945056915283, | |
| "learning_rate": 0.00012, | |
| "loss": 5.8652, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.02688486265342, | |
| "eval_loss": 5.192287445068359, | |
| "eval_runtime": 1.0975, | |
| "eval_samples_per_second": 1299.264, | |
| "eval_steps_per_second": 82.001, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.36469900642899, | |
| "eval_loss": 4.822643280029297, | |
| "eval_runtime": 1.0919, | |
| "eval_samples_per_second": 1306.028, | |
| "eval_steps_per_second": 82.428, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.70251315020456, | |
| "grad_norm": 3.0992887020111084, | |
| "learning_rate": 0.00016, | |
| "loss": 4.8065, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 18.70251315020456, | |
| "eval_loss": 4.534602165222168, | |
| "eval_runtime": 1.089, | |
| "eval_samples_per_second": 1309.442, | |
| "eval_steps_per_second": 82.644, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 21.040327293980127, | |
| "eval_loss": 4.324636459350586, | |
| "eval_runtime": 1.0927, | |
| "eval_samples_per_second": 1305.0, | |
| "eval_steps_per_second": 82.363, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 23.378141437755698, | |
| "grad_norm": 2.9212777614593506, | |
| "learning_rate": 0.0002, | |
| "loss": 4.2309, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 23.378141437755698, | |
| "eval_loss": 4.134767532348633, | |
| "eval_runtime": 1.0879, | |
| "eval_samples_per_second": 1310.811, | |
| "eval_steps_per_second": 82.73, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 25.71595558153127, | |
| "eval_loss": 3.96517276763916, | |
| "eval_runtime": 1.0898, | |
| "eval_samples_per_second": 1308.527, | |
| "eval_steps_per_second": 82.586, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 28.05376972530684, | |
| "grad_norm": 3.0659778118133545, | |
| "learning_rate": 0.00024, | |
| "loss": 3.8185, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 28.05376972530684, | |
| "eval_loss": 3.810825824737549, | |
| "eval_runtime": 1.0939, | |
| "eval_samples_per_second": 1303.648, | |
| "eval_steps_per_second": 82.278, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 30.391583869082407, | |
| "eval_loss": 3.710167646408081, | |
| "eval_runtime": 1.0929, | |
| "eval_samples_per_second": 1304.777, | |
| "eval_steps_per_second": 82.349, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 32.72939801285798, | |
| "grad_norm": 3.134622097015381, | |
| "learning_rate": 0.00028000000000000003, | |
| "loss": 3.5163, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 32.72939801285798, | |
| "eval_loss": 3.6270954608917236, | |
| "eval_runtime": 1.0898, | |
| "eval_samples_per_second": 1308.551, | |
| "eval_steps_per_second": 82.587, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 35.067212156633545, | |
| "eval_loss": 3.534950017929077, | |
| "eval_runtime": 1.0911, | |
| "eval_samples_per_second": 1306.961, | |
| "eval_steps_per_second": 82.487, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 37.40502630040912, | |
| "grad_norm": 3.2240071296691895, | |
| "learning_rate": 0.00032, | |
| "loss": 3.2957, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 37.40502630040912, | |
| "eval_loss": 3.505293607711792, | |
| "eval_runtime": 1.0901, | |
| "eval_samples_per_second": 1308.118, | |
| "eval_steps_per_second": 82.56, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 39.74284044418469, | |
| "eval_loss": 3.414364814758301, | |
| "eval_runtime": 1.1012, | |
| "eval_samples_per_second": 1294.925, | |
| "eval_steps_per_second": 81.727, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 42.080654587960254, | |
| "grad_norm": 3.1253504753112793, | |
| "learning_rate": 0.00035999999999999997, | |
| "loss": 3.1388, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 42.080654587960254, | |
| "eval_loss": 3.363224506378174, | |
| "eval_runtime": 1.1003, | |
| "eval_samples_per_second": 1296.047, | |
| "eval_steps_per_second": 81.798, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 44.41846873173583, | |
| "eval_loss": 3.3095245361328125, | |
| "eval_runtime": 1.1193, | |
| "eval_samples_per_second": 1273.985, | |
| "eval_steps_per_second": 80.406, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 46.756282875511396, | |
| "grad_norm": 2.8377697467803955, | |
| "learning_rate": 0.0004, | |
| "loss": 3.0197, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 46.756282875511396, | |
| "eval_loss": 3.3381097316741943, | |
| "eval_runtime": 1.0907, | |
| "eval_samples_per_second": 1307.437, | |
| "eval_steps_per_second": 82.517, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 49.09409701928697, | |
| "eval_loss": 3.303609609603882, | |
| "eval_runtime": 1.1029, | |
| "eval_samples_per_second": 1292.944, | |
| "eval_steps_per_second": 81.602, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 51.43191116306254, | |
| "grad_norm": 2.7318673133850098, | |
| "learning_rate": 0.00044, | |
| "loss": 2.9398, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 51.43191116306254, | |
| "eval_loss": 3.282808780670166, | |
| "eval_runtime": 1.097, | |
| "eval_samples_per_second": 1299.868, | |
| "eval_steps_per_second": 82.039, | |
| "step": 44000 | |
| } | |
| ], | |
| "logging_steps": 4000, | |
| "max_steps": 400000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 468, | |
| "save_steps": 4000, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.001 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9180204991119360.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |