| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.7540983606557377, | |
| "eval_steps": 500, | |
| "global_step": 21, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13114754098360656, | |
| "grad_norm": 38.4906907414524, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 2.4278, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.26229508196721313, | |
| "grad_norm": 39.35649873095453, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 2.4386, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.39344262295081966, | |
| "grad_norm": 15.02300885231628, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8851, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5245901639344263, | |
| "grad_norm": 30.015021153424833, | |
| "learning_rate": 1.9848077530122083e-05, | |
| "loss": 1.8762, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.6557377049180327, | |
| "grad_norm": 11.906470741405201, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 1.7922, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.7868852459016393, | |
| "grad_norm": 6.013424294278915, | |
| "learning_rate": 1.866025403784439e-05, | |
| "loss": 1.4363, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.9180327868852459, | |
| "grad_norm": 5.488571980550277, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.2963, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.9180327868852459, | |
| "eval_loss": 1.2142459154129028, | |
| "eval_runtime": 9.3088, | |
| "eval_samples_per_second": 20.626, | |
| "eval_steps_per_second": 2.578, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.0491803278688525, | |
| "grad_norm": 4.278302382312637, | |
| "learning_rate": 1.6427876096865394e-05, | |
| "loss": 1.1874, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.180327868852459, | |
| "grad_norm": 3.204774458338129, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 1.1064, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.3114754098360657, | |
| "grad_norm": 2.8914531523767013, | |
| "learning_rate": 1.342020143325669e-05, | |
| "loss": 1.0671, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4426229508196722, | |
| "grad_norm": 2.2472150773134363, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 1.0103, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.5737704918032787, | |
| "grad_norm": 2.093742944888527, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0028, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.7049180327868854, | |
| "grad_norm": 1.8675088356857141, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.9762, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.8360655737704918, | |
| "grad_norm": 1.6366608455568707, | |
| "learning_rate": 6.579798566743314e-06, | |
| "loss": 0.9678, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.9672131147540983, | |
| "grad_norm": 1.5464740113993372, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.954, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.9672131147540983, | |
| "eval_loss": 1.001825213432312, | |
| "eval_runtime": 9.0799, | |
| "eval_samples_per_second": 21.145, | |
| "eval_steps_per_second": 2.643, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.098360655737705, | |
| "grad_norm": 1.4407616891998447, | |
| "learning_rate": 3.5721239031346067e-06, | |
| "loss": 0.8805, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.2295081967213113, | |
| "grad_norm": 1.4358137476258483, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.8681, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.360655737704918, | |
| "grad_norm": 1.332633367234562, | |
| "learning_rate": 1.339745962155613e-06, | |
| "loss": 0.8592, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.4918032786885247, | |
| "grad_norm": 1.2284953779578738, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.8545, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.6229508196721314, | |
| "grad_norm": 1.2207700124089214, | |
| "learning_rate": 1.519224698779198e-07, | |
| "loss": 0.8337, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.7540983606557377, | |
| "grad_norm": 1.1315984846861744, | |
| "learning_rate": 0.0, | |
| "loss": 0.8356, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.7540983606557377, | |
| "eval_loss": 0.9876740574836731, | |
| "eval_runtime": 8.9893, | |
| "eval_samples_per_second": 21.359, | |
| "eval_steps_per_second": 2.67, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.7540983606557377, | |
| "step": 21, | |
| "total_flos": 3971544514560.0, | |
| "train_loss": 0.0, | |
| "train_runtime": 10.5592, | |
| "train_samples_per_second": 137.227, | |
| "train_steps_per_second": 1.989 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 21, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3971544514560.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |