| { | |
| "best_metric": 0.7399751659163123, | |
| "best_model_checkpoint": "/opt/dlami/nvme/shevtsov/sent_checkpoints/checkpoint-102510", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 102510, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "step": 10251, | |
| "train_accuracy": 0.9232076630758701, | |
| "train_f1": 0.9284970394554347, | |
| "train_loss": 0.08324012160301208, | |
| "train_precision": 0.9395995876182259, | |
| "train_recall": 0.9232076630758701, | |
| "train_runtime": 5077.4566, | |
| "train_samples_per_second": 775.226, | |
| "train_steps_per_second": 2.019 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.202436923980713, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1512, | |
| "step": 10251 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5906430303537085, | |
| "eval_f1": 0.5953154182610875, | |
| "eval_loss": 1.1822575330734253, | |
| "eval_precision": 0.7605817989645098, | |
| "eval_recall": 0.5906430303537085, | |
| "eval_runtime": 31.2064, | |
| "eval_samples_per_second": 766.446, | |
| "eval_steps_per_second": 2.019, | |
| "step": 10251 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 20502, | |
| "train_accuracy": 0.9251504125579815, | |
| "train_f1": 0.9318479879378413, | |
| "train_loss": 0.06029369682073593, | |
| "train_precision": 0.9457024638382566, | |
| "train_recall": 0.9251504125579815, | |
| "train_runtime": 5078.2486, | |
| "train_samples_per_second": 775.105, | |
| "train_steps_per_second": 2.019 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.3391969203948975, | |
| "learning_rate": 4.849231551964771e-06, | |
| "loss": 0.0727, | |
| "step": 20502 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5809850321933272, | |
| "eval_f1": 0.5827904211311477, | |
| "eval_loss": 0.8601031303405762, | |
| "eval_precision": 0.7707426917608643, | |
| "eval_recall": 0.5809850321933272, | |
| "eval_runtime": 31.1947, | |
| "eval_samples_per_second": 766.734, | |
| "eval_steps_per_second": 2.02, | |
| "step": 20502 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 30753, | |
| "train_accuracy": 0.9311338878819889, | |
| "train_f1": 0.9367318532276315, | |
| "train_loss": 0.05340421944856644, | |
| "train_precision": 0.948615352312463, | |
| "train_recall": 0.9311338878819889, | |
| "train_runtime": 5078.9472, | |
| "train_samples_per_second": 774.998, | |
| "train_steps_per_second": 2.018 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.368744134902954, | |
| "learning_rate": 4.415111107797445e-06, | |
| "loss": 0.064, | |
| "step": 30753 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6280207375198595, | |
| "eval_f1": 0.6341902420521265, | |
| "eval_loss": 1.0681450366973877, | |
| "eval_precision": 0.768741241964947, | |
| "eval_recall": 0.6280207375198595, | |
| "eval_runtime": 31.2028, | |
| "eval_samples_per_second": 766.535, | |
| "eval_steps_per_second": 2.019, | |
| "step": 30753 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 41004, | |
| "train_accuracy": 0.941354218588914, | |
| "train_f1": 0.9444982035100681, | |
| "train_loss": 0.0493415892124176, | |
| "train_precision": 0.9518003283876015, | |
| "train_recall": 0.941354218588914, | |
| "train_runtime": 5080.4688, | |
| "train_samples_per_second": 774.766, | |
| "train_steps_per_second": 2.018 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 7.862049579620361, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.0585, | |
| "step": 41004 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6900660590350364, | |
| "eval_f1": 0.6986283009597273, | |
| "eval_loss": 1.4099539518356323, | |
| "eval_precision": 0.7671384440115526, | |
| "eval_recall": 0.6900660590350364, | |
| "eval_runtime": 31.2008, | |
| "eval_samples_per_second": 766.584, | |
| "eval_steps_per_second": 2.019, | |
| "step": 41004 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 51255, | |
| "train_accuracy": 0.9479113474150279, | |
| "train_f1": 0.9498162368391557, | |
| "train_loss": 0.04905932769179344, | |
| "train_precision": 0.9544620214289802, | |
| "train_recall": 0.9479113474150279, | |
| "train_runtime": 5088.1804, | |
| "train_samples_per_second": 773.592, | |
| "train_steps_per_second": 2.015 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.6737421751022339, | |
| "learning_rate": 2.9341204441673267e-06, | |
| "loss": 0.0543, | |
| "step": 51255 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7279454803913371, | |
| "eval_f1": 0.7344324841621389, | |
| "eval_loss": 2.154353618621826, | |
| "eval_precision": 0.7651689935562233, | |
| "eval_recall": 0.7279454803913371, | |
| "eval_runtime": 31.2388, | |
| "eval_samples_per_second": 765.65, | |
| "eval_steps_per_second": 2.017, | |
| "step": 51255 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 61506, | |
| "train_accuracy": 0.944803507162031, | |
| "train_f1": 0.9475462857180277, | |
| "train_loss": 0.0426737517118454, | |
| "train_precision": 0.9540200071722176, | |
| "train_recall": 0.944803507162031, | |
| "train_runtime": 5078.4022, | |
| "train_samples_per_second": 775.081, | |
| "train_steps_per_second": 2.019 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.879870653152466, | |
| "learning_rate": 2.0658795558326745e-06, | |
| "loss": 0.051, | |
| "step": 61506 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7079187223011958, | |
| "eval_f1": 0.7155283465045912, | |
| "eval_loss": 2.2507946491241455, | |
| "eval_precision": 0.7570941064333973, | |
| "eval_recall": 0.7079187223011958, | |
| "eval_runtime": 31.2026, | |
| "eval_samples_per_second": 766.538, | |
| "eval_steps_per_second": 2.019, | |
| "step": 61506 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 71757, | |
| "train_accuracy": 0.9474675154096338, | |
| "train_f1": 0.9497025418465336, | |
| "train_loss": 0.0410270020365715, | |
| "train_precision": 0.9551267616744933, | |
| "train_recall": 0.9474675154096338, | |
| "train_runtime": 5080.7556, | |
| "train_samples_per_second": 774.722, | |
| "train_steps_per_second": 2.018 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.4426418542861938, | |
| "learning_rate": 1.2500000000000007e-06, | |
| "loss": 0.0486, | |
| "step": 71757 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7198344343172506, | |
| "eval_f1": 0.726834352850576, | |
| "eval_loss": 2.45487380027771, | |
| "eval_precision": 0.7590566392960729, | |
| "eval_recall": 0.7198344343172506, | |
| "eval_runtime": 31.2049, | |
| "eval_samples_per_second": 766.482, | |
| "eval_steps_per_second": 2.019, | |
| "step": 71757 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 82008, | |
| "train_accuracy": 0.9504046315025708, | |
| "train_f1": 0.9520763661297328, | |
| "train_loss": 0.0410199835896492, | |
| "train_precision": 0.9562815819971588, | |
| "train_recall": 0.9504046315025708, | |
| "train_runtime": 5079.0768, | |
| "train_samples_per_second": 774.978, | |
| "train_steps_per_second": 2.018 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.1270148754119873, | |
| "learning_rate": 5.848888922025553e-07, | |
| "loss": 0.0468, | |
| "step": 82008 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7316665273016139, | |
| "eval_f1": 0.737640105520052, | |
| "eval_loss": 2.602576494216919, | |
| "eval_precision": 0.7612609852108917, | |
| "eval_recall": 0.7316665273016139, | |
| "eval_runtime": 31.1916, | |
| "eval_samples_per_second": 766.809, | |
| "eval_steps_per_second": 2.02, | |
| "step": 82008 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "step": 92259, | |
| "train_accuracy": 0.9505296259769004, | |
| "train_f1": 0.9522083394688067, | |
| "train_loss": 0.03998752683401108, | |
| "train_precision": 0.9564247374560226, | |
| "train_recall": 0.9505296259769004, | |
| "train_runtime": 5078.0822, | |
| "train_samples_per_second": 775.13, | |
| "train_steps_per_second": 2.019 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.666390061378479, | |
| "learning_rate": 1.507684480352292e-07, | |
| "loss": 0.0457, | |
| "step": 92259 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7320010034283803, | |
| "eval_f1": 0.7378168031871946, | |
| "eval_loss": 2.8211002349853516, | |
| "eval_precision": 0.7595695441249473, | |
| "eval_recall": 0.7320010034283803, | |
| "eval_runtime": 31.2132, | |
| "eval_samples_per_second": 766.278, | |
| "eval_steps_per_second": 2.018, | |
| "step": 92259 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 102510, | |
| "train_accuracy": 0.9510209660446921, | |
| "train_f1": 0.9526073589638447, | |
| "train_loss": 0.040298543870449066, | |
| "train_precision": 0.9566123120238171, | |
| "train_recall": 0.9510209660446921, | |
| "train_runtime": 5078.6508, | |
| "train_samples_per_second": 775.043, | |
| "train_steps_per_second": 2.018 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.6141560077667236, | |
| "learning_rate": 0.0, | |
| "loss": 0.0451, | |
| "step": 102510 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7345095743791287, | |
| "eval_f1": 0.7399751659163123, | |
| "eval_loss": 2.884028911590576, | |
| "eval_precision": 0.759520061975123, | |
| "eval_recall": 0.7345095743791287, | |
| "eval_runtime": 31.2146, | |
| "eval_samples_per_second": 766.243, | |
| "eval_steps_per_second": 2.018, | |
| "step": 102510 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 102510, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0357340434899927e+19, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |