| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.6721311475409837, | |
| "eval_steps": 25, | |
| "global_step": 28, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13114754098360656, | |
| "grad_norm": 1.4051299095153809, | |
| "learning_rate": 0.00019285714285714286, | |
| "loss": 9.4338, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.26229508196721313, | |
| "grad_norm": 1.426056146621704, | |
| "learning_rate": 0.00018571428571428572, | |
| "loss": 9.3069, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.39344262295081966, | |
| "grad_norm": 2.004256248474121, | |
| "learning_rate": 0.0001785714285714286, | |
| "loss": 9.0161, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5245901639344263, | |
| "grad_norm": 2.2767462730407715, | |
| "learning_rate": 0.00017142857142857143, | |
| "loss": 8.7415, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.6557377049180327, | |
| "grad_norm": 2.5265560150146484, | |
| "learning_rate": 0.00016428571428571428, | |
| "loss": 8.2736, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.7868852459016393, | |
| "grad_norm": 2.473156690597534, | |
| "learning_rate": 0.00015714285714285716, | |
| "loss": 7.8397, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.9180327868852459, | |
| "grad_norm": 2.193089246749878, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 7.6748, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.0491803278688525, | |
| "grad_norm": 2.468278169631958, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 7.229, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.180327868852459, | |
| "grad_norm": 3.6872873306274414, | |
| "learning_rate": 0.00013571428571428572, | |
| "loss": 7.4406, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.3114754098360657, | |
| "grad_norm": 3.794747829437256, | |
| "learning_rate": 0.00012857142857142858, | |
| "loss": 7.4603, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4426229508196722, | |
| "grad_norm": 5.574512004852295, | |
| "learning_rate": 0.00012142857142857143, | |
| "loss": 7.0868, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.5737704918032787, | |
| "grad_norm": 2.9117279052734375, | |
| "learning_rate": 0.00011428571428571428, | |
| "loss": 7.14, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.7049180327868854, | |
| "grad_norm": 2.064516305923462, | |
| "learning_rate": 0.00010714285714285715, | |
| "loss": 6.8126, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.8360655737704918, | |
| "grad_norm": 1.537017583847046, | |
| "learning_rate": 0.0001, | |
| "loss": 6.8454, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.9672131147540983, | |
| "grad_norm": 1.320478081703186, | |
| "learning_rate": 9.285714285714286e-05, | |
| "loss": 6.8693, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.098360655737705, | |
| "grad_norm": 1.6444069147109985, | |
| "learning_rate": 8.571428571428571e-05, | |
| "loss": 7.0829, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.2295081967213113, | |
| "grad_norm": 2.1548044681549072, | |
| "learning_rate": 7.857142857142858e-05, | |
| "loss": 6.7618, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.360655737704918, | |
| "grad_norm": 1.982169508934021, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 6.5765, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.4918032786885247, | |
| "grad_norm": 1.1214826107025146, | |
| "learning_rate": 6.428571428571429e-05, | |
| "loss": 6.9022, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.6229508196721314, | |
| "grad_norm": 1.4072163105010986, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 6.5652, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.7540983606557377, | |
| "grad_norm": 1.1185771226882935, | |
| "learning_rate": 5e-05, | |
| "loss": 6.6532, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.8852459016393444, | |
| "grad_norm": 1.0517714023590088, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 6.7248, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.0163934426229506, | |
| "grad_norm": 1.0128535032272339, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 6.5302, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.1475409836065573, | |
| "grad_norm": 1.1242914199829102, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 6.5167, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.278688524590164, | |
| "grad_norm": 0.9026089310646057, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 6.5295, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.278688524590164, | |
| "eval_clap": 0.011066867038607597, | |
| "eval_loss": 7.892996788024902, | |
| "eval_runtime": 152.7525, | |
| "eval_samples_per_second": 0.052, | |
| "eval_steps_per_second": 0.052, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.4098360655737707, | |
| "grad_norm": 1.2630215883255005, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 6.8253, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.540983606557377, | |
| "grad_norm": 1.0664726495742798, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 6.7346, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.6721311475409837, | |
| "grad_norm": 0.969233512878418, | |
| "learning_rate": 0.0, | |
| "loss": 6.4473, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.6721311475409837, | |
| "step": 28, | |
| "total_flos": 12416744906496.0, | |
| "train_loss": 7.286445873124259, | |
| "train_runtime": 282.3834, | |
| "train_samples_per_second": 1.728, | |
| "train_steps_per_second": 0.099 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 28, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 12416744906496.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |