|
{ |
|
"best_metric": 0.3798506557941437, |
|
"best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-90", |
|
"epoch": 0.8910891089108911, |
|
"eval_steps": 10, |
|
"global_step": 90, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 0.6981944441795349, |
|
"eval_runtime": 94.6462, |
|
"eval_samples_per_second": 30.292, |
|
"eval_steps_per_second": 0.951, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"grad_norm": 0.2654201090335846, |
|
"learning_rate": 0.0001961261695938319, |
|
"loss": 1.0016, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"eval_loss": 0.467332661151886, |
|
"eval_runtime": 94.5373, |
|
"eval_samples_per_second": 30.327, |
|
"eval_steps_per_second": 0.952, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"eval_loss": 0.434685617685318, |
|
"eval_runtime": 94.5899, |
|
"eval_samples_per_second": 30.31, |
|
"eval_steps_per_second": 0.951, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 1.3763986825942993, |
|
"learning_rate": 0.0001559192903470747, |
|
"loss": 0.4357, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"eval_loss": 0.4082697033882141, |
|
"eval_runtime": 94.551, |
|
"eval_samples_per_second": 30.322, |
|
"eval_steps_per_second": 0.952, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"eval_loss": 0.40642818808555603, |
|
"eval_runtime": 94.5831, |
|
"eval_samples_per_second": 30.312, |
|
"eval_steps_per_second": 0.952, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"grad_norm": 0.16300354897975922, |
|
"learning_rate": 8.954715367323468e-05, |
|
"loss": 0.4012, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"eval_loss": 0.39164847135543823, |
|
"eval_runtime": 94.6074, |
|
"eval_samples_per_second": 30.304, |
|
"eval_steps_per_second": 0.951, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"eval_loss": 0.3856147527694702, |
|
"eval_runtime": 94.5883, |
|
"eval_samples_per_second": 30.31, |
|
"eval_steps_per_second": 0.951, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"grad_norm": 0.14875428378582, |
|
"learning_rate": 2.8066019966134904e-05, |
|
"loss": 0.3846, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"eval_loss": 0.38140159845352173, |
|
"eval_runtime": 94.5138, |
|
"eval_samples_per_second": 30.334, |
|
"eval_steps_per_second": 0.952, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"eval_loss": 0.3798506557941437, |
|
"eval_runtime": 94.6456, |
|
"eval_samples_per_second": 30.292, |
|
"eval_steps_per_second": 0.951, |
|
"step": 90 |
|
} |
|
], |
|
"logging_steps": 20, |
|
"max_steps": 100, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 10, |
|
"total_flos": 8.129633583858975e+17, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|