| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.0, |
| "eval_steps": 1000, |
| "global_step": 10116, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00019774569903104609, |
| "grad_norm": 8.964256286621094, |
| "learning_rate": 0.0, |
| "loss": 3.9486, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.19774569903104608, |
| "grad_norm": 6.0791192054748535, |
| "learning_rate": 3.2921403855659913e-06, |
| "loss": 3.172, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.19774569903104608, |
| "eval_cosine_accuracy": 0.9392154812812805, |
| "eval_loss": 0.5803369879722595, |
| "eval_runtime": 34.8035, |
| "eval_samples_per_second": 273.219, |
| "eval_steps_per_second": 2.155, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.39549139806209216, |
| "grad_norm": 6.649482250213623, |
| "learning_rate": 6.58757620695337e-06, |
| "loss": 2.6021, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.39549139806209216, |
| "eval_cosine_accuracy": 0.9489957094192505, |
| "eval_loss": 0.5286197662353516, |
| "eval_runtime": 33.7422, |
| "eval_samples_per_second": 281.813, |
| "eval_steps_per_second": 2.223, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.5932370970931382, |
| "grad_norm": 8.011848449707031, |
| "learning_rate": 9.88301202834075e-06, |
| "loss": 2.1529, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.5932370970931382, |
| "eval_cosine_accuracy": 0.9544641971588135, |
| "eval_loss": 0.49919888377189636, |
| "eval_runtime": 33.725, |
| "eval_samples_per_second": 281.957, |
| "eval_steps_per_second": 2.224, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.7909827961241843, |
| "grad_norm": 8.981022834777832, |
| "learning_rate": 1.317515241390674e-05, |
| "loss": 1.3847, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.7909827961241843, |
| "eval_cosine_accuracy": 0.9546745419502258, |
| "eval_loss": 0.47944125533103943, |
| "eval_runtime": 33.7995, |
| "eval_samples_per_second": 281.335, |
| "eval_steps_per_second": 2.219, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.9887284951552304, |
| "grad_norm": 12.642546653747559, |
| "learning_rate": 1.647058823529412e-05, |
| "loss": 0.9942, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.9887284951552304, |
| "eval_cosine_accuracy": 0.9547796845436096, |
| "eval_loss": 0.4432453513145447, |
| "eval_runtime": 33.82, |
| "eval_samples_per_second": 281.165, |
| "eval_steps_per_second": 2.218, |
| "step": 5000 |
| }, |
| { |
| "epoch": 1.1864004744020558, |
| "grad_norm": 6.489721298217773, |
| "learning_rate": 1.976272862086011e-05, |
| "loss": 1.4574, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.1864004744020558, |
| "eval_cosine_accuracy": 0.9597223401069641, |
| "eval_loss": 0.43778446316719055, |
| "eval_runtime": 35.2925, |
| "eval_samples_per_second": 269.434, |
| "eval_steps_per_second": 2.125, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.3840679976279897, |
| "grad_norm": 11.65429973602295, |
| "learning_rate": 1.923536439665472e-05, |
| "loss": 1.3286, |
| "step": 7000 |
| }, |
| { |
| "epoch": 1.3840679976279897, |
| "eval_cosine_accuracy": 0.9628772735595703, |
| "eval_loss": 0.4299427270889282, |
| "eval_runtime": 33.6219, |
| "eval_samples_per_second": 282.822, |
| "eval_steps_per_second": 2.231, |
| "step": 7000 |
| }, |
| { |
| "epoch": 1.5817355208539237, |
| "grad_norm": 6.698935508728027, |
| "learning_rate": 1.8411403617187826e-05, |
| "loss": 1.2024, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.5817355208539237, |
| "eval_cosine_accuracy": 0.9645599126815796, |
| "eval_loss": 0.41794154047966003, |
| "eval_runtime": 33.7437, |
| "eval_samples_per_second": 281.801, |
| "eval_steps_per_second": 2.223, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.7794030440798578, |
| "grad_norm": 6.8305864334106445, |
| "learning_rate": 1.758909075927986e-05, |
| "loss": 1.1554, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.7794030440798578, |
| "eval_cosine_accuracy": 0.9647701978683472, |
| "eval_loss": 0.4171123504638672, |
| "eval_runtime": 33.5248, |
| "eval_samples_per_second": 283.64, |
| "eval_steps_per_second": 2.237, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.9770705673057916, |
| "grad_norm": 6.222408771514893, |
| "learning_rate": 1.676512997981296e-05, |
| "loss": 1.0769, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.9770705673057916, |
| "eval_cosine_accuracy": 0.9635082483291626, |
| "eval_loss": 0.41735684871673584, |
| "eval_runtime": 33.8215, |
| "eval_samples_per_second": 281.152, |
| "eval_steps_per_second": 2.218, |
| "step": 10000 |
| } |
| ], |
| "logging_steps": 1000, |
| "max_steps": 30342, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 6, |
| "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": 128, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
