| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.857142857142857, | |
| "eval_steps": 50.0, | |
| "global_step": 24, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 0.14930766820907593, | |
| "learning_rate": 9.583333333333334e-05, | |
| "loss": 8.9408, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.12185841798782349, | |
| "learning_rate": 9.166666666666667e-05, | |
| "loss": 8.8871, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 0.22343124449253082, | |
| "learning_rate": 8.75e-05, | |
| "loss": 8.875, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "eval_clap": 0.067754365503788, | |
| "eval_loss": 4.407386302947998, | |
| "eval_runtime": 209.2468, | |
| "eval_samples_per_second": 0.038, | |
| "eval_steps_per_second": 0.038, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.302925169467926, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 8.8616, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.30731210112571716, | |
| "learning_rate": 7.916666666666666e-05, | |
| "loss": 8.8302, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.35163554549217224, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 8.7832, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.40824490785598755, | |
| "learning_rate": 7.083333333333334e-05, | |
| "loss": 8.7074, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_clap": 0.05861677974462509, | |
| "eval_loss": 4.424124717712402, | |
| "eval_runtime": 209.036, | |
| "eval_samples_per_second": 0.038, | |
| "eval_steps_per_second": 0.038, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.5694748759269714, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 8.6684, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.4113900065422058, | |
| "learning_rate": 6.25e-05, | |
| "loss": 8.6663, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.4741291403770447, | |
| "learning_rate": 5.833333333333334e-05, | |
| "loss": 8.6372, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_clap": 0.0703531801700592, | |
| "eval_loss": 4.461740493774414, | |
| "eval_runtime": 217.2072, | |
| "eval_samples_per_second": 0.037, | |
| "eval_steps_per_second": 0.037, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.5966173410415649, | |
| "learning_rate": 5.4166666666666664e-05, | |
| "loss": 8.5292, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.4768383204936981, | |
| "learning_rate": 5e-05, | |
| "loss": 8.5238, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.5463787913322449, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 8.5916, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5644258856773376, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 8.5145, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_clap": 0.05660417675971985, | |
| "eval_loss": 4.537668228149414, | |
| "eval_runtime": 215.7023, | |
| "eval_samples_per_second": 0.037, | |
| "eval_steps_per_second": 0.037, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.4694964587688446, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 8.446, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.5955868363380432, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 8.5089, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.5509549975395203, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 8.4215, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_clap": 0.056698545813560486, | |
| "eval_loss": 4.5900163650512695, | |
| "eval_runtime": 209.8722, | |
| "eval_samples_per_second": 0.038, | |
| "eval_steps_per_second": 0.038, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 0.6947008967399597, | |
| "learning_rate": 2.5e-05, | |
| "loss": 8.4253, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 0.609464168548584, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 8.4624, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.49679213762283325, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 8.3189, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.6397719383239746, | |
| "learning_rate": 1.25e-05, | |
| "loss": 8.4121, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_clap": 0.03204277530312538, | |
| "eval_loss": 4.638527870178223, | |
| "eval_runtime": 216.9058, | |
| "eval_samples_per_second": 0.037, | |
| "eval_steps_per_second": 0.037, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.6107072830200195, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 8.3785, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 0.553840696811676, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 8.3048, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "grad_norm": 0.5590672492980957, | |
| "learning_rate": 0.0, | |
| "loss": 8.377, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_clap": 0.053874481469392776, | |
| "eval_loss": 4.658874988555908, | |
| "eval_runtime": 216.879, | |
| "eval_samples_per_second": 0.037, | |
| "eval_steps_per_second": 0.037, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "step": 24, | |
| "total_flos": 22951666441296.0, | |
| "train_loss": 8.58632747332255, | |
| "train_runtime": 1809.0113, | |
| "train_samples_per_second": 0.239, | |
| "train_steps_per_second": 0.013 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 24, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "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": 22951666441296.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |