| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "eval_steps": 4350, | |
| "global_step": 43500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.00021520511654671282, | |
| "learning_rate": 9.000919540229886e-06, | |
| "loss": 0.48, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 3.651794195175171, | |
| "eval_runtime": 55.7113, | |
| "eval_samples_per_second": 311.66, | |
| "eval_steps_per_second": 7.09, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.00017535020015202463, | |
| "learning_rate": 8.001379310344829e-06, | |
| "loss": 0.1193, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 3.453643798828125, | |
| "eval_runtime": 55.7389, | |
| "eval_samples_per_second": 311.506, | |
| "eval_steps_per_second": 7.087, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.18407990038394928, | |
| "learning_rate": 7.00183908045977e-06, | |
| "loss": 0.0925, | |
| "step": 13050 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 3.4169740676879883, | |
| "eval_runtime": 55.9578, | |
| "eval_samples_per_second": 310.287, | |
| "eval_steps_per_second": 7.059, | |
| "step": 13050 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 8.286705269711092e-05, | |
| "learning_rate": 6.002298850574713e-06, | |
| "loss": 0.0827, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 3.3203234672546387, | |
| "eval_runtime": 55.9217, | |
| "eval_samples_per_second": 310.488, | |
| "eval_steps_per_second": 7.063, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.0003959204477723688, | |
| "learning_rate": 5.002758620689656e-06, | |
| "loss": 0.0755, | |
| "step": 21750 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 3.196115732192993, | |
| "eval_runtime": 55.4125, | |
| "eval_samples_per_second": 313.341, | |
| "eval_steps_per_second": 7.128, | |
| "step": 21750 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 7.149695011321455e-05, | |
| "learning_rate": 4.003218390804598e-06, | |
| "loss": 0.0685, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 3.110957145690918, | |
| "eval_runtime": 55.9729, | |
| "eval_samples_per_second": 310.204, | |
| "eval_steps_per_second": 7.057, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 0.013322222046554089, | |
| "learning_rate": 3.0036781609195404e-06, | |
| "loss": 0.0664, | |
| "step": 30450 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_loss": 3.0240492820739746, | |
| "eval_runtime": 56.1145, | |
| "eval_samples_per_second": 309.421, | |
| "eval_steps_per_second": 7.039, | |
| "step": 30450 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 2.9004069801885635e-05, | |
| "learning_rate": 2.004137931034483e-06, | |
| "loss": 0.0621, | |
| "step": 34800 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_loss": 2.9804065227508545, | |
| "eval_runtime": 55.7487, | |
| "eval_samples_per_second": 311.451, | |
| "eval_steps_per_second": 7.085, | |
| "step": 34800 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 0.00013569927250500768, | |
| "learning_rate": 1.0045977011494254e-06, | |
| "loss": 0.0596, | |
| "step": 39150 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_loss": 2.895709276199341, | |
| "eval_runtime": 55.8728, | |
| "eval_samples_per_second": 310.76, | |
| "eval_steps_per_second": 7.07, | |
| "step": 39150 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 0.0004280584107618779, | |
| "learning_rate": 4.827586206896552e-09, | |
| "loss": 0.0576, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 2.854401111602783, | |
| "eval_runtime": 55.466, | |
| "eval_samples_per_second": 313.038, | |
| "eval_steps_per_second": 7.121, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 43500, | |
| "total_flos": 6.859464121840128e+17, | |
| "train_loss": 0.11642176906541846, | |
| "train_runtime": 15360.6557, | |
| "train_samples_per_second": 113.036, | |
| "train_steps_per_second": 2.832 | |
| } | |
| ], | |
| "logging_steps": 4350, | |
| "max_steps": 43500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "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": 6.859464121840128e+17, | |
| "train_batch_size": 40, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |