| { | |
| "best_metric": 1.2563587427139282, | |
| "best_model_checkpoint": "finetuned-bert__java-8m-methods__jemma-java-bert-tokenizer/checkpoint-5000", | |
| "epoch": 4.669020454611771, | |
| "global_step": 40000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0004888116319438376, | |
| "loss": 3.0057, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.2563587427139282, | |
| "eval_runtime": 836.9828, | |
| "eval_samples_per_second": 59.343, | |
| "eval_steps_per_second": 7.418, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00044517243817696517, | |
| "loss": 1.5788, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 6.417298793792725, | |
| "eval_runtime": 835.6884, | |
| "eval_samples_per_second": 59.435, | |
| "eval_steps_per_second": 7.43, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00037428394451934934, | |
| "loss": 5.9109, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 6.413046836853027, | |
| "eval_runtime": 833.6127, | |
| "eval_samples_per_second": 59.583, | |
| "eval_steps_per_second": 7.448, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.00028607867802092244, | |
| "loss": 5.9036, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 6.433701038360596, | |
| "eval_runtime": 833.0303, | |
| "eval_samples_per_second": 59.624, | |
| "eval_steps_per_second": 7.454, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00019284648488204127, | |
| "loss": 5.9003, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 6.432703971862793, | |
| "eval_runtime": 833.1905, | |
| "eval_samples_per_second": 59.613, | |
| "eval_steps_per_second": 7.452, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.00010757762475595697, | |
| "loss": 5.8971, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_loss": 6.4252238273620605, | |
| "eval_runtime": 846.7073, | |
| "eval_samples_per_second": 58.661, | |
| "eval_steps_per_second": 7.333, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 4.217367458660157e-05, | |
| "loss": 5.8972, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "eval_loss": 6.429380416870117, | |
| "eval_runtime": 846.9722, | |
| "eval_samples_per_second": 58.643, | |
| "eval_steps_per_second": 7.331, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 5.6997776508484205e-06, | |
| "loss": 5.892, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_loss": 6.426671504974365, | |
| "eval_runtime": 847.0538, | |
| "eval_samples_per_second": 58.637, | |
| "eval_steps_per_second": 7.33, | |
| "step": 40000 | |
| } | |
| ], | |
| "max_steps": 42835, | |
| "num_train_epochs": 5, | |
| "total_flos": 6.741181575431455e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |