| { | |
| "best_metric": 0.6586877107620239, | |
| "best_model_checkpoint": "saves/Gemma-7B/lora/train_1/checkpoint-100", | |
| "epoch": 0.21875854525567404, | |
| "eval_steps": 10, | |
| "global_step": 200, | |
| "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": 0.7906885147094727, | |
| "eval_runtime": 170.0021, | |
| "eval_samples_per_second": 27.717, | |
| "eval_steps_per_second": 0.871, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021875854525567406, | |
| "grad_norm": 1.8665393590927124, | |
| "learning_rate": 0.00029996062078124905, | |
| "loss": 0.6717, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021875854525567406, | |
| "eval_loss": 0.7608615159988403, | |
| "eval_runtime": 168.7111, | |
| "eval_samples_per_second": 27.929, | |
| "eval_steps_per_second": 0.877, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03281378178835111, | |
| "grad_norm": 2.046764373779297, | |
| "learning_rate": 0.0002999114016039678, | |
| "loss": 0.5741, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03281378178835111, | |
| "eval_loss": 0.7403788566589355, | |
| "eval_runtime": 169.3669, | |
| "eval_samples_per_second": 27.821, | |
| "eval_steps_per_second": 0.874, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04375170905113481, | |
| "grad_norm": 0.8763797879219055, | |
| "learning_rate": 0.00029984250380130117, | |
| "loss": 0.5809, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04375170905113481, | |
| "eval_loss": 0.7739446759223938, | |
| "eval_runtime": 169.4581, | |
| "eval_samples_per_second": 27.806, | |
| "eval_steps_per_second": 0.873, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05468963631391851, | |
| "grad_norm": 1.452701449394226, | |
| "learning_rate": 0.0002997539364173515, | |
| "loss": 0.5313, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05468963631391851, | |
| "eval_loss": 0.7001951932907104, | |
| "eval_runtime": 169.6448, | |
| "eval_samples_per_second": 27.776, | |
| "eval_steps_per_second": 0.872, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06562756357670221, | |
| "grad_norm": 2.0603904724121094, | |
| "learning_rate": 0.00029964571107821494, | |
| "loss": 0.4879, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06562756357670221, | |
| "eval_loss": 0.7159304022789001, | |
| "eval_runtime": 169.4832, | |
| "eval_samples_per_second": 27.802, | |
| "eval_steps_per_second": 0.873, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07656549083948591, | |
| "grad_norm": 1.048985242843628, | |
| "learning_rate": 0.00029951784199045534, | |
| "loss": 0.4665, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.07656549083948591, | |
| "eval_loss": 0.7063425779342651, | |
| "eval_runtime": 169.1768, | |
| "eval_samples_per_second": 27.853, | |
| "eval_steps_per_second": 0.875, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08750341810226962, | |
| "grad_norm": 0.8847436904907227, | |
| "learning_rate": 0.0002993703459392396, | |
| "loss": 0.4509, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.08750341810226962, | |
| "eval_loss": 0.6991614103317261, | |
| "eval_runtime": 169.2261, | |
| "eval_samples_per_second": 27.844, | |
| "eval_steps_per_second": 0.875, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09844134536505332, | |
| "grad_norm": 1.09526789188385, | |
| "learning_rate": 0.00029920324228613376, | |
| "loss": 0.4542, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.09844134536505332, | |
| "eval_loss": 0.6915357708930969, | |
| "eval_runtime": 169.6804, | |
| "eval_samples_per_second": 27.77, | |
| "eval_steps_per_second": 0.872, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10937927262783702, | |
| "grad_norm": 0.9263470768928528, | |
| "learning_rate": 0.0002990165529665622, | |
| "loss": 0.4188, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.10937927262783702, | |
| "eval_loss": 0.6586877107620239, | |
| "eval_runtime": 169.6821, | |
| "eval_samples_per_second": 27.77, | |
| "eval_steps_per_second": 0.872, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12031719989062073, | |
| "grad_norm": 1.2452396154403687, | |
| "learning_rate": 0.0002988103024869277, | |
| "loss": 0.4131, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12031719989062073, | |
| "eval_loss": 0.6637363433837891, | |
| "eval_runtime": 169.6991, | |
| "eval_samples_per_second": 27.767, | |
| "eval_steps_per_second": 0.872, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13125512715340443, | |
| "grad_norm": 0.8261873722076416, | |
| "learning_rate": 0.00029858451792139453, | |
| "loss": 0.4137, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13125512715340443, | |
| "eval_loss": 0.6902170181274414, | |
| "eval_runtime": 169.6732, | |
| "eval_samples_per_second": 27.771, | |
| "eval_steps_per_second": 0.872, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.14219305441618812, | |
| "grad_norm": 0.6665583848953247, | |
| "learning_rate": 0.0002983392289083346, | |
| "loss": 0.4087, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14219305441618812, | |
| "eval_loss": 0.6948944330215454, | |
| "eval_runtime": 169.5204, | |
| "eval_samples_per_second": 27.796, | |
| "eval_steps_per_second": 0.873, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.15313098167897182, | |
| "grad_norm": 0.841665506362915, | |
| "learning_rate": 0.0002980744676464371, | |
| "loss": 0.3968, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15313098167897182, | |
| "eval_loss": 0.6712561845779419, | |
| "eval_runtime": 169.6453, | |
| "eval_samples_per_second": 27.776, | |
| "eval_steps_per_second": 0.872, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.16406890894175555, | |
| "grad_norm": 1.1116673946380615, | |
| "learning_rate": 0.0002977902688904813, | |
| "loss": 0.4048, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16406890894175555, | |
| "eval_loss": 0.6878468990325928, | |
| "eval_runtime": 169.3026, | |
| "eval_samples_per_second": 27.832, | |
| "eval_steps_per_second": 0.874, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "grad_norm": 0.7878606915473938, | |
| "learning_rate": 0.00029748666994677467, | |
| "loss": 0.3953, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17500683620453925, | |
| "eval_loss": 0.6906653642654419, | |
| "eval_runtime": 168.3921, | |
| "eval_samples_per_second": 27.982, | |
| "eval_steps_per_second": 0.879, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.18594476346732294, | |
| "grad_norm": 0.9028112888336182, | |
| "learning_rate": 0.00029716371066825593, | |
| "loss": 0.3873, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18594476346732294, | |
| "eval_loss": 0.6937726140022278, | |
| "eval_runtime": 169.1341, | |
| "eval_samples_per_second": 27.86, | |
| "eval_steps_per_second": 0.875, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.19688269073010664, | |
| "grad_norm": 0.797646701335907, | |
| "learning_rate": 0.0002968214334492632, | |
| "loss": 0.3821, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19688269073010664, | |
| "eval_loss": 0.684823751449585, | |
| "eval_runtime": 169.2007, | |
| "eval_samples_per_second": 27.849, | |
| "eval_steps_per_second": 0.875, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.20782061799289034, | |
| "grad_norm": 0.7943875193595886, | |
| "learning_rate": 0.00029645988321996917, | |
| "loss": 0.394, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.20782061799289034, | |
| "eval_loss": 0.7038875222206116, | |
| "eval_runtime": 169.1579, | |
| "eval_samples_per_second": 27.856, | |
| "eval_steps_per_second": 0.875, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.21875854525567404, | |
| "grad_norm": 0.797392725944519, | |
| "learning_rate": 0.00029607910744048336, | |
| "loss": 0.3893, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21875854525567404, | |
| "eval_loss": 0.6831381916999817, | |
| "eval_runtime": 169.2814, | |
| "eval_samples_per_second": 27.835, | |
| "eval_steps_per_second": 0.874, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21875854525567404, | |
| "step": 200, | |
| "total_flos": 3.6702510649442304e+17, | |
| "train_loss": 0.5889949607849121, | |
| "train_runtime": 10495.9143, | |
| "train_samples_per_second": 66.895, | |
| "train_steps_per_second": 0.261 | |
| } | |
| ], | |
| "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": 10, | |
| "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": 3.6702510649442304e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |