| { | |
| "best_metric": 0.6592887043952942, | |
| "best_model_checkpoint": "saves/Gemma-7B/lora/train_1/checkpoint-110", | |
| "epoch": 0.17500683620453925, | |
| "eval_steps": 10, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010937927262783703, | |
| "grad_norm": 16.269868850708008, | |
| "learning_rate": 0.00029999015487222375, | |
| "loss": 3.1586, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.010937927262783703, | |
| "eval_loss": 0.7969412207603455, | |
| "eval_runtime": 1611.4779, | |
| "eval_samples_per_second": 29.24, | |
| "eval_steps_per_second": 0.914, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021875854525567406, | |
| "grad_norm": 1.8665393590927124, | |
| "learning_rate": 0.00029996062078124905, | |
| "loss": 0.6717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021875854525567406, | |
| "eval_loss": 0.7616342306137085, | |
| "eval_runtime": 1610.0951, | |
| "eval_samples_per_second": 29.265, | |
| "eval_steps_per_second": 0.915, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03281378178835111, | |
| "grad_norm": 2.046764373779297, | |
| "learning_rate": 0.0002999114016039678, | |
| "loss": 0.5741, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03281378178835111, | |
| "eval_loss": 0.7397875189781189, | |
| "eval_runtime": 1609.4859, | |
| "eval_samples_per_second": 29.276, | |
| "eval_steps_per_second": 0.915, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04375170905113481, | |
| "grad_norm": 0.8763797879219055, | |
| "learning_rate": 0.00029984250380130117, | |
| "loss": 0.5809, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04375170905113481, | |
| "eval_loss": 0.7744876742362976, | |
| "eval_runtime": 1610.4604, | |
| "eval_samples_per_second": 29.259, | |
| "eval_steps_per_second": 0.915, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05468963631391851, | |
| "grad_norm": 1.452701449394226, | |
| "learning_rate": 0.0002997539364173515, | |
| "loss": 0.5313, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05468963631391851, | |
| "eval_loss": 0.7013862729072571, | |
| "eval_runtime": 1609.7654, | |
| "eval_samples_per_second": 29.271, | |
| "eval_steps_per_second": 0.915, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06562756357670221, | |
| "grad_norm": 2.0603904724121094, | |
| "learning_rate": 0.00029964571107821494, | |
| "loss": 0.4879, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06562756357670221, | |
| "eval_loss": 0.7164642214775085, | |
| "eval_runtime": 1609.8109, | |
| "eval_samples_per_second": 29.271, | |
| "eval_steps_per_second": 0.915, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07656549083948591, | |
| "grad_norm": 1.048985242843628, | |
| "learning_rate": 0.00029951784199045534, | |
| "loss": 0.4665, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.07656549083948591, | |
| "eval_loss": 0.7070000171661377, | |
| "eval_runtime": 1608.8038, | |
| "eval_samples_per_second": 29.289, | |
| "eval_steps_per_second": 0.916, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08750341810226962, | |
| "grad_norm": 0.8847436904907227, | |
| "learning_rate": 0.0002993703459392396, | |
| "loss": 0.4509, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.08750341810226962, | |
| "eval_loss": 0.7005062699317932, | |
| "eval_runtime": 1608.5117, | |
| "eval_samples_per_second": 29.294, | |
| "eval_steps_per_second": 0.916, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09844134536505332, | |
| "grad_norm": 1.09526789188385, | |
| "learning_rate": 0.00029920324228613376, | |
| "loss": 0.4542, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.09844134536505332, | |
| "eval_loss": 0.6932889223098755, | |
| "eval_runtime": 1608.5476, | |
| "eval_samples_per_second": 29.294, | |
| "eval_steps_per_second": 0.916, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10937927262783702, | |
| "grad_norm": 0.9263470768928528, | |
| "learning_rate": 0.0002990165529665622, | |
| "loss": 0.4188, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.10937927262783702, | |
| "eval_loss": 0.65990149974823, | |
| "eval_runtime": 1607.8192, | |
| "eval_samples_per_second": 29.307, | |
| "eval_steps_per_second": 0.916, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12031719989062073, | |
| "grad_norm": 1.2452396154403687, | |
| "learning_rate": 0.0002988103024869277, | |
| "loss": 0.4131, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12031719989062073, | |
| "eval_loss": 0.6592887043952942, | |
| "eval_runtime": 1608.173, | |
| "eval_samples_per_second": 29.3, | |
| "eval_steps_per_second": 0.916, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13125512715340443, | |
| "grad_norm": 0.8261873722076416, | |
| "learning_rate": 0.00029858451792139453, | |
| "loss": 0.4137, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13125512715340443, | |
| "eval_loss": 0.6895950436592102, | |
| "eval_runtime": 1609.8507, | |
| "eval_samples_per_second": 29.27, | |
| "eval_steps_per_second": 0.915, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.14219305441618812, | |
| "grad_norm": 0.6665583848953247, | |
| "learning_rate": 0.0002983392289083346, | |
| "loss": 0.4087, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14219305441618812, | |
| "eval_loss": 0.6930198073387146, | |
| "eval_runtime": 1610.0518, | |
| "eval_samples_per_second": 29.266, | |
| "eval_steps_per_second": 0.915, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.15313098167897182, | |
| "grad_norm": 0.841665506362915, | |
| "learning_rate": 0.0002980744676464371, | |
| "loss": 0.3968, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15313098167897182, | |
| "eval_loss": 0.6691634058952332, | |
| "eval_runtime": 1610.3191, | |
| "eval_samples_per_second": 29.261, | |
| "eval_steps_per_second": 0.915, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.16406890894175555, | |
| "grad_norm": 1.1116673946380615, | |
| "learning_rate": 0.0002977902688904813, | |
| "loss": 0.4048, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16406890894175555, | |
| "eval_loss": 0.6872161030769348, | |
| "eval_runtime": 1608.6952, | |
| "eval_samples_per_second": 29.291, | |
| "eval_steps_per_second": 0.916, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "grad_norm": 0.7878606915473938, | |
| "learning_rate": 0.00029748666994677467, | |
| "loss": 0.3953, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "eval_loss": 0.6938689351081848, | |
| "eval_runtime": 1607.7411, | |
| "eval_samples_per_second": 29.308, | |
| "eval_steps_per_second": 0.916, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "step": 160, | |
| "total_flos": 2.934122540040192e+17, | |
| "train_loss": 0.639202207326889, | |
| "train_runtime": 31420.6578, | |
| "train_samples_per_second": 22.346, | |
| "train_steps_per_second": 0.087 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2742, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.934122540040192e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |