| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.008233910267182, | |
| "eval_steps": 500, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08401949252226516, | |
| "grad_norm": 19.443201065063477, | |
| "learning_rate": 0.0002455, | |
| "loss": 4.6277, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08401949252226516, | |
| "eval_loss": 1.7780689001083374, | |
| "eval_runtime": 184.9372, | |
| "eval_samples_per_second": 37.97, | |
| "eval_steps_per_second": 2.374, | |
| "eval_wer": 0.9507301452317996, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.16803898504453033, | |
| "grad_norm": 10.621150970458984, | |
| "learning_rate": 0.0002782777777777778, | |
| "loss": 1.2514, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.16803898504453033, | |
| "eval_loss": 1.046749472618103, | |
| "eval_runtime": 183.9043, | |
| "eval_samples_per_second": 38.183, | |
| "eval_steps_per_second": 2.387, | |
| "eval_wer": 0.7116623925377695, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2520584775667955, | |
| "grad_norm": 6.651599884033203, | |
| "learning_rate": 0.00025049999999999996, | |
| "loss": 0.986, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2520584775667955, | |
| "eval_loss": 0.8940034508705139, | |
| "eval_runtime": 185.3076, | |
| "eval_samples_per_second": 37.894, | |
| "eval_steps_per_second": 2.369, | |
| "eval_wer": 0.6577817935396431, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.33607797008906065, | |
| "grad_norm": 16.908123016357422, | |
| "learning_rate": 0.00022277777777777774, | |
| "loss": 0.8702, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.33607797008906065, | |
| "eval_loss": 0.8157631158828735, | |
| "eval_runtime": 185.5186, | |
| "eval_samples_per_second": 37.851, | |
| "eval_steps_per_second": 2.366, | |
| "eval_wer": 0.5836644122297665, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4200974626113258, | |
| "grad_norm": 10.515896797180176, | |
| "learning_rate": 0.00019505555555555555, | |
| "loss": 0.8063, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.4200974626113258, | |
| "eval_loss": 0.7131509184837341, | |
| "eval_runtime": 185.8391, | |
| "eval_samples_per_second": 37.785, | |
| "eval_steps_per_second": 2.362, | |
| "eval_wer": 0.5281095948657304, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.504116955133591, | |
| "grad_norm": 3.7379024028778076, | |
| "learning_rate": 0.00016727777777777778, | |
| "loss": 0.7518, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.504116955133591, | |
| "eval_loss": 0.7195897102355957, | |
| "eval_runtime": 183.5821, | |
| "eval_samples_per_second": 38.25, | |
| "eval_steps_per_second": 2.391, | |
| "eval_wer": 0.5187950942744389, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5881364476558562, | |
| "grad_norm": 5.922393321990967, | |
| "learning_rate": 0.0001395, | |
| "loss": 0.7045, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5881364476558562, | |
| "eval_loss": 0.6231045722961426, | |
| "eval_runtime": 184.2635, | |
| "eval_samples_per_second": 38.108, | |
| "eval_steps_per_second": 2.382, | |
| "eval_wer": 0.4771256593895747, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6721559401781213, | |
| "grad_norm": 8.91370964050293, | |
| "learning_rate": 0.00011177777777777777, | |
| "loss": 0.6726, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6721559401781213, | |
| "eval_loss": 0.6074743270874023, | |
| "eval_runtime": 184.3908, | |
| "eval_samples_per_second": 38.082, | |
| "eval_steps_per_second": 2.381, | |
| "eval_wer": 0.4517333475066105, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7561754327003865, | |
| "grad_norm": 12.337292671203613, | |
| "learning_rate": 8.405555555555556e-05, | |
| "loss": 0.6355, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.7561754327003865, | |
| "eval_loss": 0.5601897239685059, | |
| "eval_runtime": 184.3021, | |
| "eval_samples_per_second": 38.1, | |
| "eval_steps_per_second": 2.382, | |
| "eval_wer": 0.4328784597191034, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.8401949252226516, | |
| "grad_norm": 18.518247604370117, | |
| "learning_rate": 5.6333333333333325e-05, | |
| "loss": 0.5938, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.8401949252226516, | |
| "eval_loss": 0.5438016653060913, | |
| "eval_runtime": 185.1047, | |
| "eval_samples_per_second": 37.935, | |
| "eval_steps_per_second": 2.372, | |
| "eval_wer": 0.4225939754713722, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.9242144177449169, | |
| "grad_norm": 5.622241020202637, | |
| "learning_rate": 2.8555555555555556e-05, | |
| "loss": 0.5842, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.9242144177449169, | |
| "eval_loss": 0.5245193839073181, | |
| "eval_runtime": 184.9427, | |
| "eval_samples_per_second": 37.969, | |
| "eval_steps_per_second": 2.374, | |
| "eval_wer": 0.4090939289653065, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.008233910267182, | |
| "grad_norm": 10.519143104553223, | |
| "learning_rate": 7.777777777777777e-07, | |
| "loss": 0.557, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.008233910267182, | |
| "eval_loss": 0.5155353546142578, | |
| "eval_runtime": 185.543, | |
| "eval_samples_per_second": 37.846, | |
| "eval_steps_per_second": 2.366, | |
| "eval_wer": 0.39834438406037814, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.008233910267182, | |
| "step": 6000, | |
| "total_flos": 6.665403914161851e+18, | |
| "train_loss": 1.086754165649414, | |
| "train_runtime": 4718.794, | |
| "train_samples_per_second": 10.172, | |
| "train_steps_per_second": 1.272 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 6.665403914161851e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |