| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.584507042253521, | |
| "eval_steps": 500, | |
| "global_step": 1800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0880281690140845, | |
| "grad_norm": 1221.827392578125, | |
| "learning_rate": 2.7777777777777783e-06, | |
| "loss": 297.1973, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.176056338028169, | |
| "grad_norm": 1063.7159423828125, | |
| "learning_rate": 4.998119881260576e-06, | |
| "loss": 161.9498, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2640845070422535, | |
| "grad_norm": 1415.1790771484375, | |
| "learning_rate": 4.93261217644956e-06, | |
| "loss": 149.2985, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.352112676056338, | |
| "grad_norm": 1143.650634765625, | |
| "learning_rate": 4.775907352415367e-06, | |
| "loss": 146.5141, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.44014084507042256, | |
| "grad_norm": 1112.0836181640625, | |
| "learning_rate": 4.533880175657419e-06, | |
| "loss": 125.0505, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.528169014084507, | |
| "grad_norm": 2647.099365234375, | |
| "learning_rate": 4.215604094671835e-06, | |
| "loss": 136.0549, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6161971830985915, | |
| "grad_norm": 914.672119140625, | |
| "learning_rate": 3.833011082004229e-06, | |
| "loss": 152.9934, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.704225352112676, | |
| "grad_norm": 941.3635864257812, | |
| "learning_rate": 3.400444312011776e-06, | |
| "loss": 142.6481, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7922535211267606, | |
| "grad_norm": 2750.364990234375, | |
| "learning_rate": 2.9341204441673267e-06, | |
| "loss": 180.6238, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.8802816901408451, | |
| "grad_norm": 1102.6456298828125, | |
| "learning_rate": 2.4515216705704396e-06, | |
| "loss": 160.7811, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9683098591549296, | |
| "grad_norm": 2127.451416015625, | |
| "learning_rate": 1.970740319426474e-06, | |
| "loss": 136.9254, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 415.93927001953125, | |
| "eval_runtime": 8.7817, | |
| "eval_samples_per_second": 115.012, | |
| "eval_steps_per_second": 14.462, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 1.056338028169014, | |
| "grad_norm": 2116.73193359375, | |
| "learning_rate": 1.509800584902108e-06, | |
| "loss": 142.92, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.1443661971830985, | |
| "grad_norm": 1667.754638671875, | |
| "learning_rate": 1.085982811283654e-06, | |
| "loss": 141.8554, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.232394366197183, | |
| "grad_norm": 1366.0006103515625, | |
| "learning_rate": 7.151756636052529e-07, | |
| "loss": 149.8122, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.3204225352112675, | |
| "grad_norm": 1805.1434326171875, | |
| "learning_rate": 4.1128047146765936e-07, | |
| "loss": 148.2215, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.408450704225352, | |
| "grad_norm": 1603.68212890625, | |
| "learning_rate": 1.8569007682777417e-07, | |
| "loss": 142.1948, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.4964788732394365, | |
| "grad_norm": 1250.8704833984375, | |
| "learning_rate": 4.6861723431538273e-08, | |
| "loss": 123.6116, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.584507042253521, | |
| "grad_norm": 3193.298828125, | |
| "learning_rate": 0.0, | |
| "loss": 142.5536, | |
| "step": 1800 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 600, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |