accumulate_grad_batches: 1
args_path: YOUR_EXPERIMENT_DIR\dualcse_mono-model=bert-bs=16-lr=1-loss=3\args.yaml
batch_size: 16
best_ckpt_path: YOUR_EXPERIMENT_DIR\dualcse_mono-model=bert-bs=16-lr=1-loss=3\global_step=2700.pt
best_score: 0.765749990940094
check_val_every_n_epoch: null
experiment_dir: YOUR_EXPERIMENT_DIR\dualcse_mono-model=bert-bs=16-lr=1-loss=3
gpus: 1
log_every_n_steps: 10
loss: 3
lr: 1.0e-05
max_epochs: 10
max_seq_len: 64
method: dualcse_mono
model_name: bert-base-uncased
num_nodes: 1
precision: 32
profiler: false
run_name: dualcse_mono-model=bert-bs=16-lr=1-loss=3
strategy: auto
task_name: rte
temperature: 0.05
val_check_interval: 100
wandb: true