| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 2000, |
| "global_step": 23091, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00012992074834351045, |
| "grad_norm": 7.7499589920043945, |
| "learning_rate": 0.0, |
| "loss": 3.4107, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.2598414966870209, |
| "grad_norm": 3.223598003387451, |
| "learning_rate": 7.792359667359668e-06, |
| "loss": 2.4514, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.2598414966870209, |
| "eval_cosine_accuracy": 0.946060299873352, |
| "eval_loss": 0.91331946849823, |
| "eval_runtime": 25.9781, |
| "eval_samples_per_second": 363.961, |
| "eval_steps_per_second": 0.731, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.5196829933740418, |
| "grad_norm": 2.9365298748016357, |
| "learning_rate": 1.5588617463617465e-05, |
| "loss": 1.6277, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.5196829933740418, |
| "eval_cosine_accuracy": 0.9472236633300781, |
| "eval_loss": 0.8556821942329407, |
| "eval_runtime": 25.551, |
| "eval_samples_per_second": 370.044, |
| "eval_steps_per_second": 0.744, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.7795244900610627, |
| "grad_norm": 3.0075325965881348, |
| "learning_rate": 2.338097713097713e-05, |
| "loss": 1.2469, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.7795244900610627, |
| "eval_cosine_accuracy": 0.9501850605010986, |
| "eval_loss": 0.8617205619812012, |
| "eval_runtime": 25.8371, |
| "eval_samples_per_second": 365.947, |
| "eval_steps_per_second": 0.735, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.0393659867480838, |
| "grad_norm": 2.5449373722076416, |
| "learning_rate": 2.9987508340635427e-05, |
| "loss": 1.2862, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.0393659867480838, |
| "eval_cosine_accuracy": 0.9508196711540222, |
| "eval_loss": 0.8610702157020569, |
| "eval_runtime": 25.794, |
| "eval_samples_per_second": 366.559, |
| "eval_steps_per_second": 0.737, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.2992074834351046, |
| "grad_norm": 2.4445102214813232, |
| "learning_rate": 2.927163819523492e-05, |
| "loss": 0.9041, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.2992074834351046, |
| "eval_cosine_accuracy": 0.956213653087616, |
| "eval_loss": 0.7799679636955261, |
| "eval_runtime": 25.5138, |
| "eval_samples_per_second": 370.584, |
| "eval_steps_per_second": 0.745, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.5590489801221254, |
| "grad_norm": 2.134432077407837, |
| "learning_rate": 2.7505671097607588e-05, |
| "loss": 0.8181, |
| "step": 12000 |
| }, |
| { |
| "epoch": 1.5590489801221254, |
| "eval_cosine_accuracy": 0.9569540023803711, |
| "eval_loss": 0.7651481628417969, |
| "eval_runtime": 25.6352, |
| "eval_samples_per_second": 368.829, |
| "eval_steps_per_second": 0.741, |
| "step": 12000 |
| }, |
| { |
| "epoch": 1.8188904768091465, |
| "grad_norm": 2.1797378063201904, |
| "learning_rate": 2.4821088211734495e-05, |
| "loss": 0.7655, |
| "step": 14000 |
| }, |
| { |
| "epoch": 1.8188904768091465, |
| "eval_cosine_accuracy": 0.962136447429657, |
| "eval_loss": 0.7638042569160461, |
| "eval_runtime": 25.5173, |
| "eval_samples_per_second": 370.532, |
| "eval_steps_per_second": 0.745, |
| "step": 14000 |
| }, |
| { |
| "epoch": 2.0787319734961676, |
| "grad_norm": 2.1507601737976074, |
| "learning_rate": 2.1416439286393234e-05, |
| "loss": 0.99, |
| "step": 16000 |
| }, |
| { |
| "epoch": 2.0787319734961676, |
| "eval_cosine_accuracy": 0.9612902998924255, |
| "eval_loss": 0.7620325088500977, |
| "eval_runtime": 25.8187, |
| "eval_samples_per_second": 366.208, |
| "eval_steps_per_second": 0.736, |
| "step": 16000 |
| }, |
| { |
| "epoch": 2.338573470183188, |
| "grad_norm": 2.0006659030914307, |
| "learning_rate": 1.7536588752614515e-05, |
| "loss": 0.6912, |
| "step": 18000 |
| }, |
| { |
| "epoch": 2.338573470183188, |
| "eval_cosine_accuracy": 0.9635113477706909, |
| "eval_loss": 0.745087742805481, |
| "eval_runtime": 25.6534, |
| "eval_samples_per_second": 368.567, |
| "eval_steps_per_second": 0.741, |
| "step": 18000 |
| }, |
| { |
| "epoch": 2.5984149668702092, |
| "grad_norm": 1.920875072479248, |
| "learning_rate": 1.3470097828872529e-05, |
| "loss": 0.671, |
| "step": 20000 |
| }, |
| { |
| "epoch": 2.5984149668702092, |
| "eval_cosine_accuracy": 0.9632998704910278, |
| "eval_loss": 0.724320113658905, |
| "eval_runtime": 25.7545, |
| "eval_samples_per_second": 367.121, |
| "eval_steps_per_second": 0.738, |
| "step": 20000 |
| }, |
| { |
| "epoch": 2.85825646355723, |
| "grad_norm": 1.8667397499084473, |
| "learning_rate": 9.516176014936201e-06, |
| "loss": 0.6589, |
| "step": 22000 |
| }, |
| { |
| "epoch": 2.85825646355723, |
| "eval_cosine_accuracy": 0.9643574953079224, |
| "eval_loss": 0.7122445702552795, |
| "eval_runtime": 25.4494, |
| "eval_samples_per_second": 371.521, |
| "eval_steps_per_second": 0.747, |
| "step": 22000 |
| } |
| ], |
| "logging_steps": 2000, |
| "max_steps": 30788, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 500, |
| "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": 512, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
