| { | |
| "best_metric": 1.4656140804290771, | |
| "best_model_checkpoint": "models/dehanalkautsar/mbert-3-with-multilingual-tokenizer-30k/checkpoint-24000", | |
| "epoch": 9.996235629260353, | |
| "eval_steps": 2000, | |
| "global_step": 24570, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8139179977617255, | |
| "grad_norm": 18.654558181762695, | |
| "learning_rate": 9.185999185999187e-05, | |
| "loss": 4.0579, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8139179977617255, | |
| "eval_loss": 2.5225632190704346, | |
| "eval_runtime": 73.2103, | |
| "eval_samples_per_second": 136.593, | |
| "eval_steps_per_second": 2.145, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.6275307762742903, | |
| "grad_norm": 15.436388969421387, | |
| "learning_rate": 8.371998371998372e-05, | |
| "loss": 2.3568, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6275307762742903, | |
| "eval_loss": 2.068246603012085, | |
| "eval_runtime": 73.2939, | |
| "eval_samples_per_second": 136.437, | |
| "eval_steps_per_second": 2.142, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.4411435547868554, | |
| "grad_norm": 13.966560363769531, | |
| "learning_rate": 7.557997557997558e-05, | |
| "loss": 2.0445, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.4411435547868554, | |
| "eval_loss": 1.8596677780151367, | |
| "eval_runtime": 73.1178, | |
| "eval_samples_per_second": 136.766, | |
| "eval_steps_per_second": 2.147, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.25475633329942, | |
| "grad_norm": 14.865880012512207, | |
| "learning_rate": 6.743996743996744e-05, | |
| "loss": 1.8887, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.25475633329942, | |
| "eval_loss": 1.7477023601531982, | |
| "eval_runtime": 73.1164, | |
| "eval_samples_per_second": 136.768, | |
| "eval_steps_per_second": 2.147, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.068369111811985, | |
| "grad_norm": 13.934277534484863, | |
| "learning_rate": 5.929995929995931e-05, | |
| "loss": 1.7863, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.068369111811985, | |
| "eval_loss": 1.677494764328003, | |
| "eval_runtime": 73.1043, | |
| "eval_samples_per_second": 136.791, | |
| "eval_steps_per_second": 2.148, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.882287109573711, | |
| "grad_norm": 13.954177856445312, | |
| "learning_rate": 5.115995115995116e-05, | |
| "loss": 1.7168, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.882287109573711, | |
| "eval_loss": 1.6307039260864258, | |
| "eval_runtime": 73.09, | |
| "eval_samples_per_second": 136.818, | |
| "eval_steps_per_second": 2.148, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.6958998880862755, | |
| "grad_norm": 13.182450294494629, | |
| "learning_rate": 4.301994301994302e-05, | |
| "loss": 1.6626, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 5.6958998880862755, | |
| "eval_loss": 1.5682964324951172, | |
| "eval_runtime": 73.1095, | |
| "eval_samples_per_second": 136.781, | |
| "eval_steps_per_second": 2.147, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.50951266659884, | |
| "grad_norm": 13.514598846435547, | |
| "learning_rate": 3.487993487993488e-05, | |
| "loss": 1.6187, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.50951266659884, | |
| "eval_loss": 1.5308398008346558, | |
| "eval_runtime": 73.125, | |
| "eval_samples_per_second": 136.752, | |
| "eval_steps_per_second": 2.147, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 7.323125445111405, | |
| "grad_norm": 14.169295310974121, | |
| "learning_rate": 2.673992673992674e-05, | |
| "loss": 1.5873, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.323125445111405, | |
| "eval_loss": 1.5098525285720825, | |
| "eval_runtime": 73.107, | |
| "eval_samples_per_second": 136.786, | |
| "eval_steps_per_second": 2.148, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 8.13673822362397, | |
| "grad_norm": 13.590775489807129, | |
| "learning_rate": 1.85999185999186e-05, | |
| "loss": 1.5568, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.13673822362397, | |
| "eval_loss": 1.4912240505218506, | |
| "eval_runtime": 73.2463, | |
| "eval_samples_per_second": 136.526, | |
| "eval_steps_per_second": 2.143, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.950656221385696, | |
| "grad_norm": 13.897253036499023, | |
| "learning_rate": 1.045991045991046e-05, | |
| "loss": 1.5348, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 8.950656221385696, | |
| "eval_loss": 1.4697929620742798, | |
| "eval_runtime": 73.2504, | |
| "eval_samples_per_second": 136.518, | |
| "eval_steps_per_second": 2.143, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.76426899989826, | |
| "grad_norm": 13.867361068725586, | |
| "learning_rate": 2.31990231990232e-06, | |
| "loss": 1.5172, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 9.76426899989826, | |
| "eval_loss": 1.4656140804290771, | |
| "eval_runtime": 73.2214, | |
| "eval_samples_per_second": 136.572, | |
| "eval_steps_per_second": 2.144, | |
| "step": 24000 | |
| } | |
| ], | |
| "logging_steps": 2000, | |
| "max_steps": 24570, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 2000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.6550302836575232e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |