| { | |
| "best_metric": 0.6494305729866028, | |
| "best_model_checkpoint": "llama2_short/checkpoint-550", | |
| "epoch": 5.238095238095238, | |
| "eval_steps": 25, | |
| "global_step": 550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23809523809523808, | |
| "grad_norm": 0.09585642069578171, | |
| "learning_rate": 9.986018985905901e-05, | |
| "loss": 0.8756, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.23809523809523808, | |
| "eval_loss": 0.8215582370758057, | |
| "eval_runtime": 31.8561, | |
| "eval_samples_per_second": 3.484, | |
| "eval_steps_per_second": 0.439, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 0.09733700007200241, | |
| "learning_rate": 9.944154131125642e-05, | |
| "loss": 0.789, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "eval_loss": 0.7655627131462097, | |
| "eval_runtime": 31.8837, | |
| "eval_samples_per_second": 3.481, | |
| "eval_steps_per_second": 0.439, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 0.11074452102184296, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.759, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "eval_loss": 0.7467875480651855, | |
| "eval_runtime": 31.8741, | |
| "eval_samples_per_second": 3.482, | |
| "eval_steps_per_second": 0.439, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 0.10002691298723221, | |
| "learning_rate": 9.777864028930705e-05, | |
| "loss": 0.7402, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "eval_loss": 0.7353290915489197, | |
| "eval_runtime": 31.8697, | |
| "eval_samples_per_second": 3.483, | |
| "eval_steps_per_second": 0.439, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1904761904761905, | |
| "grad_norm": 0.12271334230899811, | |
| "learning_rate": 9.654368743221022e-05, | |
| "loss": 0.7267, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.1904761904761905, | |
| "eval_loss": 0.7255786657333374, | |
| "eval_runtime": 31.8838, | |
| "eval_samples_per_second": 3.481, | |
| "eval_steps_per_second": 0.439, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.15203851461410522, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.7228, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "eval_loss": 0.7192813754081726, | |
| "eval_runtime": 31.8619, | |
| "eval_samples_per_second": 3.484, | |
| "eval_steps_per_second": 0.439, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.1312723308801651, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.7173, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "eval_loss": 0.7122732400894165, | |
| "eval_runtime": 31.8673, | |
| "eval_samples_per_second": 3.483, | |
| "eval_steps_per_second": 0.439, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.1437642127275467, | |
| "learning_rate": 9.131193871579975e-05, | |
| "loss": 0.7194, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "eval_loss": 0.7064856886863708, | |
| "eval_runtime": 31.8744, | |
| "eval_samples_per_second": 3.482, | |
| "eval_steps_per_second": 0.439, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 0.16143007576465607, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.6987, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "eval_loss": 0.7001689076423645, | |
| "eval_runtime": 31.8599, | |
| "eval_samples_per_second": 3.484, | |
| "eval_steps_per_second": 0.439, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.1549287736415863, | |
| "learning_rate": 8.665259359149132e-05, | |
| "loss": 0.6941, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "eval_loss": 0.6950200200080872, | |
| "eval_runtime": 31.9601, | |
| "eval_samples_per_second": 3.473, | |
| "eval_steps_per_second": 0.438, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.619047619047619, | |
| "grad_norm": 0.15042544901371002, | |
| "learning_rate": 8.400863688854597e-05, | |
| "loss": 0.6987, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.619047619047619, | |
| "eval_loss": 0.6901896595954895, | |
| "eval_runtime": 31.9342, | |
| "eval_samples_per_second": 3.476, | |
| "eval_steps_per_second": 0.438, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.17507144808769226, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.6953, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_loss": 0.6860001087188721, | |
| "eval_runtime": 31.8997, | |
| "eval_samples_per_second": 3.48, | |
| "eval_steps_per_second": 0.439, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0952380952380953, | |
| "grad_norm": 0.16489411890506744, | |
| "learning_rate": 7.81660029031811e-05, | |
| "loss": 0.6832, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 3.0952380952380953, | |
| "eval_loss": 0.681731104850769, | |
| "eval_runtime": 31.8876, | |
| "eval_samples_per_second": 3.481, | |
| "eval_steps_per_second": 0.439, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.17146223783493042, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.6784, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "eval_loss": 0.6773594617843628, | |
| "eval_runtime": 31.8931, | |
| "eval_samples_per_second": 3.48, | |
| "eval_steps_per_second": 0.439, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 0.1854155957698822, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.6756, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "eval_loss": 0.6735982298851013, | |
| "eval_runtime": 31.966, | |
| "eval_samples_per_second": 3.472, | |
| "eval_steps_per_second": 0.438, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.19128429889678955, | |
| "learning_rate": 6.826705121831976e-05, | |
| "loss": 0.6771, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "eval_loss": 0.66991126537323, | |
| "eval_runtime": 31.8972, | |
| "eval_samples_per_second": 3.48, | |
| "eval_steps_per_second": 0.439, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0476190476190474, | |
| "grad_norm": 0.2203032523393631, | |
| "learning_rate": 6.473775872054521e-05, | |
| "loss": 0.6761, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 4.0476190476190474, | |
| "eval_loss": 0.6663544774055481, | |
| "eval_runtime": 31.9622, | |
| "eval_samples_per_second": 3.473, | |
| "eval_steps_per_second": 0.438, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.2110411375761032, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.664, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "eval_loss": 0.6635292768478394, | |
| "eval_runtime": 31.9093, | |
| "eval_samples_per_second": 3.479, | |
| "eval_steps_per_second": 0.439, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.523809523809524, | |
| "grad_norm": 0.21833977103233337, | |
| "learning_rate": 5.745211330880872e-05, | |
| "loss": 0.6618, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 4.523809523809524, | |
| "eval_loss": 0.6595950722694397, | |
| "eval_runtime": 31.9849, | |
| "eval_samples_per_second": 3.47, | |
| "eval_steps_per_second": 0.438, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.21036215126514435, | |
| "learning_rate": 5.373650467932122e-05, | |
| "loss": 0.6614, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "eval_loss": 0.6555557250976562, | |
| "eval_runtime": 31.9033, | |
| "eval_samples_per_second": 3.479, | |
| "eval_steps_per_second": 0.439, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.2142489105463028, | |
| "learning_rate": 5e-05, | |
| "loss": 0.662, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6540645360946655, | |
| "eval_runtime": 31.9092, | |
| "eval_samples_per_second": 3.479, | |
| "eval_steps_per_second": 0.439, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "grad_norm": 0.22267910838127136, | |
| "learning_rate": 4.626349532067879e-05, | |
| "loss": 0.6497, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.238095238095238, | |
| "eval_loss": 0.6494305729866028, | |
| "eval_runtime": 31.9204, | |
| "eval_samples_per_second": 3.477, | |
| "eval_steps_per_second": 0.439, | |
| "step": 550 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1050, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 25, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.5814765698954035e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |