| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.847457627118644, | |
| "eval_steps": 50, | |
| "global_step": 750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05649717514124294, | |
| "grad_norm": 12.016854286193848, | |
| "learning_rate": 1.0786516853932584e-05, | |
| "loss": 0.7484, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05649717514124294, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_dot_accuracy": 0.01272264631043257, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9821882951653944, | |
| "eval_all-nli-dev_max_accuracy": 0.9872773536895675, | |
| "eval_loss": 0.25870630145072937, | |
| "eval_runtime": 13.2055, | |
| "eval_samples_per_second": 29.76, | |
| "eval_steps_per_second": 3.786, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11299435028248588, | |
| "grad_norm": 16.70142936706543, | |
| "learning_rate": 1.977386934673367e-05, | |
| "loss": 0.2822, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11299435028248588, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_dot_accuracy": 0.01272264631043257, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_max_accuracy": 0.989821882951654, | |
| "eval_loss": 0.23130479454994202, | |
| "eval_runtime": 13.1695, | |
| "eval_samples_per_second": 29.842, | |
| "eval_steps_per_second": 3.797, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1694915254237288, | |
| "grad_norm": 5.665423393249512, | |
| "learning_rate": 1.8542713567839195e-05, | |
| "loss": 0.3023, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.1694915254237288, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_dot_accuracy": 0.015267175572519083, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_max_accuracy": 0.9872773536895675, | |
| "eval_loss": 0.22914756834506989, | |
| "eval_runtime": 13.1614, | |
| "eval_samples_per_second": 29.86, | |
| "eval_steps_per_second": 3.799, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.22598870056497175, | |
| "grad_norm": 0.9494842886924744, | |
| "learning_rate": 1.7311557788944723e-05, | |
| "loss": 0.2484, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22598870056497175, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_dot_accuracy": 0.01272264631043257, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_max_accuracy": 0.9872773536895675, | |
| "eval_loss": 0.21549050509929657, | |
| "eval_runtime": 13.4649, | |
| "eval_samples_per_second": 29.187, | |
| "eval_steps_per_second": 3.713, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2824858757062147, | |
| "grad_norm": 5.140994071960449, | |
| "learning_rate": 1.6055276381909547e-05, | |
| "loss": 0.2909, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2824858757062147, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_dot_accuracy": 0.015267175572519083, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9821882951653944, | |
| "eval_all-nli-dev_max_accuracy": 0.9847328244274809, | |
| "eval_loss": 0.19652578234672546, | |
| "eval_runtime": 13.5616, | |
| "eval_samples_per_second": 28.979, | |
| "eval_steps_per_second": 3.687, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.3389830508474576, | |
| "grad_norm": 18.279523849487305, | |
| "learning_rate": 1.4824120603015077e-05, | |
| "loss": 0.2999, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3389830508474576, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_dot_accuracy": 0.015267175572519083, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_max_accuracy": 0.9847328244274809, | |
| "eval_loss": 0.20084014534950256, | |
| "eval_runtime": 13.2009, | |
| "eval_samples_per_second": 29.771, | |
| "eval_steps_per_second": 3.788, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3954802259887006, | |
| "grad_norm": 4.213184833526611, | |
| "learning_rate": 1.3567839195979901e-05, | |
| "loss": 0.2586, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.3954802259887006, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9923664122137404, | |
| "eval_all-nli-dev_dot_accuracy": 0.007633587786259542, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.9923664122137404, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9923664122137404, | |
| "eval_all-nli-dev_max_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.16702787578105927, | |
| "eval_runtime": 13.3509, | |
| "eval_samples_per_second": 29.436, | |
| "eval_steps_per_second": 3.745, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4519774011299435, | |
| "grad_norm": 30.387929916381836, | |
| "learning_rate": 1.2336683417085429e-05, | |
| "loss": 0.2385, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4519774011299435, | |
| "eval_all-nli-dev_cosine_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_dot_accuracy": 0.010178117048346057, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_max_accuracy": 0.989821882951654, | |
| "eval_loss": 0.14671088755130768, | |
| "eval_runtime": 13.2819, | |
| "eval_samples_per_second": 29.589, | |
| "eval_steps_per_second": 3.765, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5084745762711864, | |
| "grad_norm": 3.245051860809326, | |
| "learning_rate": 1.1080402010050253e-05, | |
| "loss": 0.2353, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5084745762711864, | |
| "eval_all-nli-dev_cosine_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_dot_accuracy": 0.010178117048346057, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_max_accuracy": 0.989821882951654, | |
| "eval_loss": 0.13109469413757324, | |
| "eval_runtime": 13.4569, | |
| "eval_samples_per_second": 29.204, | |
| "eval_steps_per_second": 3.716, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5649717514124294, | |
| "grad_norm": 32.116214752197266, | |
| "learning_rate": 9.824120603015075e-06, | |
| "loss": 0.2632, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5649717514124294, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_dot_accuracy": 0.015267175572519083, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.9847328244274809, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_max_accuracy": 0.9872773536895675, | |
| "eval_loss": 0.13404284417629242, | |
| "eval_runtime": 13.1995, | |
| "eval_samples_per_second": 29.774, | |
| "eval_steps_per_second": 3.788, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6214689265536724, | |
| "grad_norm": 33.70884704589844, | |
| "learning_rate": 8.5678391959799e-06, | |
| "loss": 0.3793, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6214689265536724, | |
| "eval_all-nli-dev_cosine_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_dot_accuracy": 0.010178117048346057, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_max_accuracy": 0.989821882951654, | |
| "eval_loss": 0.12181754410266876, | |
| "eval_runtime": 13.194, | |
| "eval_samples_per_second": 29.786, | |
| "eval_steps_per_second": 3.79, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6779661016949152, | |
| "grad_norm": 3.5105509757995605, | |
| "learning_rate": 7.311557788944724e-06, | |
| "loss": 0.1978, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6779661016949152, | |
| "eval_all-nli-dev_cosine_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_dot_accuracy": 0.01272264631043257, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_max_accuracy": 0.989821882951654, | |
| "eval_loss": 0.11738275736570358, | |
| "eval_runtime": 13.3865, | |
| "eval_samples_per_second": 29.358, | |
| "eval_steps_per_second": 3.735, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7344632768361582, | |
| "grad_norm": 8.318052291870117, | |
| "learning_rate": 6.055276381909548e-06, | |
| "loss": 0.179, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.7344632768361582, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_dot_accuracy": 0.01272264631043257, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.9872773536895675, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_max_accuracy": 0.989821882951654, | |
| "eval_loss": 0.12535005807876587, | |
| "eval_runtime": 13.465, | |
| "eval_samples_per_second": 29.187, | |
| "eval_steps_per_second": 3.713, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.7909604519774012, | |
| "grad_norm": 26.912717819213867, | |
| "learning_rate": 4.7989949748743725e-06, | |
| "loss": 0.1326, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7909604519774012, | |
| "eval_all-nli-dev_cosine_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_dot_accuracy": 0.010178117048346057, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9923664122137404, | |
| "eval_all-nli-dev_max_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.11420778185129166, | |
| "eval_runtime": 13.2565, | |
| "eval_samples_per_second": 29.646, | |
| "eval_steps_per_second": 3.772, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.847457627118644, | |
| "grad_norm": 12.022055625915527, | |
| "learning_rate": 3.542713567839196e-06, | |
| "loss": 0.1842, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.847457627118644, | |
| "eval_all-nli-dev_cosine_accuracy": 0.9923664122137404, | |
| "eval_all-nli-dev_dot_accuracy": 0.007633587786259542, | |
| "eval_all-nli-dev_euclidean_accuracy": 0.989821882951654, | |
| "eval_all-nli-dev_manhattan_accuracy": 0.9923664122137404, | |
| "eval_all-nli-dev_max_accuracy": 0.9923664122137404, | |
| "eval_loss": 0.11530788987874985, | |
| "eval_runtime": 13.2795, | |
| "eval_samples_per_second": 29.595, | |
| "eval_steps_per_second": 3.765, | |
| "step": 750 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 885, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |