| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 3000, | |
| "global_step": 44343, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.76544212164265e-05, | |
| "grad_norm": 6.0, | |
| "learning_rate": 0.0002999932345578783, | |
| "loss": 4.375, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.20296326364927947, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 0.000279703673635072, | |
| "loss": 2.7717, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.20296326364927947, | |
| "eval_loss": 2.3463752269744873, | |
| "eval_runtime": 61.2837, | |
| "eval_samples_per_second": 1540.458, | |
| "eval_steps_per_second": 6.021, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.40592652729855894, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 0.0002594073472701441, | |
| "loss": 2.6786, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.40592652729855894, | |
| "eval_loss": 2.3242292404174805, | |
| "eval_runtime": 61.2847, | |
| "eval_samples_per_second": 1540.433, | |
| "eval_steps_per_second": 6.021, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6088897909478385, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 0.0002391110209052161, | |
| "loss": 2.6556, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.6088897909478385, | |
| "eval_loss": 2.3117165565490723, | |
| "eval_runtime": 61.1596, | |
| "eval_samples_per_second": 1543.584, | |
| "eval_steps_per_second": 6.033, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.8118530545971179, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 0.0002188146945402882, | |
| "loss": 2.6474, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.8118530545971179, | |
| "eval_loss": 2.30464506149292, | |
| "eval_runtime": 61.272, | |
| "eval_samples_per_second": 1540.754, | |
| "eval_steps_per_second": 6.022, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0148163182463974, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 0.00019851836817536025, | |
| "loss": 2.6412, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.0148163182463974, | |
| "eval_loss": 2.302506685256958, | |
| "eval_runtime": 61.3123, | |
| "eval_samples_per_second": 1539.74, | |
| "eval_steps_per_second": 6.018, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.217779581895677, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 0.0001782220418104323, | |
| "loss": 2.6352, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.217779581895677, | |
| "eval_loss": 2.299394369125366, | |
| "eval_runtime": 61.2885, | |
| "eval_samples_per_second": 1540.338, | |
| "eval_steps_per_second": 6.021, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.4207428455449564, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 0.00015792571544550436, | |
| "loss": 2.6334, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.4207428455449564, | |
| "eval_loss": 2.2973618507385254, | |
| "eval_runtime": 61.2764, | |
| "eval_samples_per_second": 1540.642, | |
| "eval_steps_per_second": 6.022, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.6237061091942357, | |
| "grad_norm": 0.8515625, | |
| "learning_rate": 0.0001376293890805764, | |
| "loss": 2.6313, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.6237061091942357, | |
| "eval_loss": 2.2973830699920654, | |
| "eval_runtime": 61.1563, | |
| "eval_samples_per_second": 1543.668, | |
| "eval_steps_per_second": 6.034, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.8266693728435153, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 0.00011733306271564845, | |
| "loss": 2.629, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.8266693728435153, | |
| "eval_loss": 2.297150135040283, | |
| "eval_runtime": 61.2918, | |
| "eval_samples_per_second": 1540.254, | |
| "eval_steps_per_second": 6.02, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.029632636492795, | |
| "grad_norm": 0.8671875, | |
| "learning_rate": 9.703673635072052e-05, | |
| "loss": 2.6292, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.029632636492795, | |
| "eval_loss": 2.2963457107543945, | |
| "eval_runtime": 61.4261, | |
| "eval_samples_per_second": 1536.887, | |
| "eval_steps_per_second": 6.007, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.232595900142074, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 7.674040998579256e-05, | |
| "loss": 2.6262, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.232595900142074, | |
| "eval_loss": 2.2962822914123535, | |
| "eval_runtime": 61.2609, | |
| "eval_samples_per_second": 1541.031, | |
| "eval_steps_per_second": 6.023, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.435559163791354, | |
| "grad_norm": 0.9921875, | |
| "learning_rate": 5.644408362086462e-05, | |
| "loss": 2.6296, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.435559163791354, | |
| "eval_loss": 2.2963669300079346, | |
| "eval_runtime": 61.2798, | |
| "eval_samples_per_second": 1540.557, | |
| "eval_steps_per_second": 6.022, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.638522427440633, | |
| "grad_norm": 0.94140625, | |
| "learning_rate": 3.614775725593667e-05, | |
| "loss": 2.63, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.638522427440633, | |
| "eval_loss": 2.2954142093658447, | |
| "eval_runtime": 61.3072, | |
| "eval_samples_per_second": 1539.869, | |
| "eval_steps_per_second": 6.019, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.841485691089913, | |
| "grad_norm": 0.88671875, | |
| "learning_rate": 1.5851430891008727e-05, | |
| "loss": 2.6285, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.841485691089913, | |
| "eval_loss": 2.295731782913208, | |
| "eval_runtime": 61.1502, | |
| "eval_samples_per_second": 1543.823, | |
| "eval_steps_per_second": 6.034, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 44343, | |
| "total_flos": 4.6159075948098355e+17, | |
| "train_loss": 2.646649062422479, | |
| "train_runtime": 22762.4415, | |
| "train_samples_per_second": 498.693, | |
| "train_steps_per_second": 1.948 | |
| } | |
| ], | |
| "logging_steps": 3000, | |
| "max_steps": 44343, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 3000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.6159075948098355e+17, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |