| { | |
| "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": 1.2684494256973267, | |
| "learning_rate": 9.583333333333334e-05, | |
| "loss": 9.5898, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 1.1818389892578125, | |
| "learning_rate": 9.166666666666667e-05, | |
| "loss": 9.5011, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 1.2637732028961182, | |
| "learning_rate": 8.75e-05, | |
| "loss": 9.4039, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "eval_clap": -0.006993885152041912, | |
| "eval_loss": 4.442792892456055, | |
| "eval_runtime": 152.7878, | |
| "eval_samples_per_second": 0.052, | |
| "eval_steps_per_second": 0.052, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.3919429779052734, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 9.2841, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.372936487197876, | |
| "learning_rate": 7.916666666666666e-05, | |
| "loss": 9.1687, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 1.4058455228805542, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 9.0566, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.5498394966125488, | |
| "learning_rate": 7.083333333333334e-05, | |
| "loss": 8.8992, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_clap": 0.011688459664583206, | |
| "eval_loss": 4.4434919357299805, | |
| "eval_runtime": 159.5674, | |
| "eval_samples_per_second": 0.05, | |
| "eval_steps_per_second": 0.05, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 1.9276645183563232, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 8.7901, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 1.75799560546875, | |
| "learning_rate": 6.25e-05, | |
| "loss": 8.7254, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 2.2315149307250977, | |
| "learning_rate": 5.833333333333334e-05, | |
| "loss": 8.5373, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_clap": -0.02875398099422455, | |
| "eval_loss": 4.446320533752441, | |
| "eval_runtime": 162.9065, | |
| "eval_samples_per_second": 0.049, | |
| "eval_steps_per_second": 0.049, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 2.4357290267944336, | |
| "learning_rate": 5.4166666666666664e-05, | |
| "loss": 8.4189, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 2.5207338333129883, | |
| "learning_rate": 5e-05, | |
| "loss": 8.2969, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 3.3077168464660645, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 8.2279, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.170753240585327, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 8.0105, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_clap": -0.011095928959548473, | |
| "eval_loss": 4.455233573913574, | |
| "eval_runtime": 156.9405, | |
| "eval_samples_per_second": 0.051, | |
| "eval_steps_per_second": 0.051, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 2.024324417114258, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 7.9972, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 1.7727833986282349, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 7.9157, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 1.7100613117218018, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 7.8068, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_clap": -0.019424768164753914, | |
| "eval_loss": 4.468198776245117, | |
| "eval_runtime": 150.9644, | |
| "eval_samples_per_second": 0.053, | |
| "eval_steps_per_second": 0.053, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 1.5242968797683716, | |
| "learning_rate": 2.5e-05, | |
| "loss": 7.7691, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 1.3554269075393677, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 7.7686, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 1.5143285989761353, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 7.7224, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.2512257099151611, | |
| "learning_rate": 1.25e-05, | |
| "loss": 7.6881, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_clap": 0.007096702232956886, | |
| "eval_loss": 4.4912824630737305, | |
| "eval_runtime": 151.9292, | |
| "eval_samples_per_second": 0.053, | |
| "eval_steps_per_second": 0.053, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 1.1607741117477417, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 7.6731, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 1.4850616455078125, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 7.6376, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "grad_norm": 1.2155146598815918, | |
| "learning_rate": 0.0, | |
| "loss": 7.6672, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_clap": -0.02221839316189289, | |
| "eval_loss": 4.501054763793945, | |
| "eval_runtime": 152.1978, | |
| "eval_samples_per_second": 0.053, | |
| "eval_steps_per_second": 0.053, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "step": 24, | |
| "total_flos": 10310332824144.0, | |
| "train_loss": 8.39816782871882, | |
| "train_runtime": 1346.8351, | |
| "train_samples_per_second": 0.321, | |
| "train_steps_per_second": 0.018 | |
| } | |
| ], | |
| "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": 10310332824144.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |