| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 226, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 1.3860198259353638, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.4089, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 1.945064663887024, | |
| "learning_rate": 7.6e-05, | |
| "loss": 1.2541, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 1.5190799236297607, | |
| "learning_rate": 0.000116, | |
| "loss": 1.0915, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 1.624822735786438, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 1.1231, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 1.3709652423858643, | |
| "learning_rate": 0.000196, | |
| "loss": 1.0294, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 1.4600133895874023, | |
| "learning_rate": 0.00018977272727272728, | |
| "loss": 1.0663, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6222222222222222, | |
| "grad_norm": 1.397749662399292, | |
| "learning_rate": 0.00017840909090909092, | |
| "loss": 1.0412, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 1.1549971103668213, | |
| "learning_rate": 0.00016704545454545455, | |
| "loss": 0.9761, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.229239583015442, | |
| "learning_rate": 0.00015568181818181818, | |
| "loss": 1.0562, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 1.3344461917877197, | |
| "learning_rate": 0.00014431818181818182, | |
| "loss": 1.0214, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "grad_norm": 1.0738836526870728, | |
| "learning_rate": 0.00013295454545454548, | |
| "loss": 0.9623, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.0325431823730469, | |
| "eval_runtime": 6.5587, | |
| "eval_samples_per_second": 15.247, | |
| "eval_steps_per_second": 1.982, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 1.0622222222222222, | |
| "grad_norm": 1.6436502933502197, | |
| "learning_rate": 0.00012159090909090908, | |
| "loss": 0.8405, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.1511111111111112, | |
| "grad_norm": 1.2771726846694946, | |
| "learning_rate": 0.00011022727272727273, | |
| "loss": 0.8454, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 1.2871211767196655, | |
| "learning_rate": 9.886363636363637e-05, | |
| "loss": 0.8083, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.3288888888888888, | |
| "grad_norm": 1.5843398571014404, | |
| "learning_rate": 8.75e-05, | |
| "loss": 0.8218, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.4177777777777778, | |
| "grad_norm": 1.2667443752288818, | |
| "learning_rate": 7.613636363636363e-05, | |
| "loss": 0.8261, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.5066666666666668, | |
| "grad_norm": 1.3391046524047852, | |
| "learning_rate": 6.477272727272728e-05, | |
| "loss": 0.8019, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.5955555555555554, | |
| "grad_norm": 1.2577234506607056, | |
| "learning_rate": 5.340909090909091e-05, | |
| "loss": 0.8598, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6844444444444444, | |
| "grad_norm": 1.5575358867645264, | |
| "learning_rate": 4.204545454545455e-05, | |
| "loss": 0.768, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7733333333333334, | |
| "grad_norm": 1.369371771812439, | |
| "learning_rate": 3.068181818181818e-05, | |
| "loss": 0.8109, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8622222222222222, | |
| "grad_norm": 1.4712281227111816, | |
| "learning_rate": 1.9318181818181818e-05, | |
| "loss": 0.7924, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.951111111111111, | |
| "grad_norm": 1.3420695066452026, | |
| "learning_rate": 7.954545454545455e-06, | |
| "loss": 0.7288, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.0223336219787598, | |
| "eval_runtime": 6.5358, | |
| "eval_samples_per_second": 15.3, | |
| "eval_steps_per_second": 1.989, | |
| "step": 226 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 226, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 5.92047891283968e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |