| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.963855421686747, | |
| "eval_steps": 10, | |
| "global_step": 186, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1606425702811245, | |
| "grad_norm": 8.837871551513672, | |
| "learning_rate": 0.000189247311827957, | |
| "loss": 8.4808, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1606425702811245, | |
| "eval_loss": 0.47449949383735657, | |
| "eval_runtime": 44.2521, | |
| "eval_samples_per_second": 11.276, | |
| "eval_steps_per_second": 2.825, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.321285140562249, | |
| "grad_norm": 3.9745054244995117, | |
| "learning_rate": 0.00017849462365591398, | |
| "loss": 3.0872, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.321285140562249, | |
| "eval_loss": 0.3382565677165985, | |
| "eval_runtime": 48.3781, | |
| "eval_samples_per_second": 10.315, | |
| "eval_steps_per_second": 2.584, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4819277108433735, | |
| "grad_norm": 3.2844247817993164, | |
| "learning_rate": 0.00016774193548387098, | |
| "loss": 2.7654, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4819277108433735, | |
| "eval_loss": 0.3172107934951782, | |
| "eval_runtime": 50.1957, | |
| "eval_samples_per_second": 9.941, | |
| "eval_steps_per_second": 2.49, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.642570281124498, | |
| "grad_norm": 2.986614465713501, | |
| "learning_rate": 0.00015698924731182796, | |
| "loss": 2.4167, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.642570281124498, | |
| "eval_loss": 0.3065057694911957, | |
| "eval_runtime": 49.9091, | |
| "eval_samples_per_second": 9.998, | |
| "eval_steps_per_second": 2.505, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8032128514056225, | |
| "grad_norm": 2.47356915473938, | |
| "learning_rate": 0.00014623655913978496, | |
| "loss": 2.5307, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8032128514056225, | |
| "eval_loss": 0.296438992023468, | |
| "eval_runtime": 45.7506, | |
| "eval_samples_per_second": 10.907, | |
| "eval_steps_per_second": 2.732, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.963855421686747, | |
| "grad_norm": 2.9974865913391113, | |
| "learning_rate": 0.00013548387096774193, | |
| "loss": 2.48, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.963855421686747, | |
| "eval_loss": 0.2858414649963379, | |
| "eval_runtime": 47.1672, | |
| "eval_samples_per_second": 10.579, | |
| "eval_steps_per_second": 2.65, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1124497991967872, | |
| "grad_norm": 2.8883159160614014, | |
| "learning_rate": 0.00012473118279569893, | |
| "loss": 1.9267, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1124497991967872, | |
| "eval_loss": 0.28903448581695557, | |
| "eval_runtime": 45.0342, | |
| "eval_samples_per_second": 11.08, | |
| "eval_steps_per_second": 2.776, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2730923694779117, | |
| "grad_norm": 2.6527364253997803, | |
| "learning_rate": 0.00011397849462365593, | |
| "loss": 1.9188, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.2730923694779117, | |
| "eval_loss": 0.284964382648468, | |
| "eval_runtime": 44.0349, | |
| "eval_samples_per_second": 11.332, | |
| "eval_steps_per_second": 2.839, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.4337349397590362, | |
| "grad_norm": 3.009997606277466, | |
| "learning_rate": 0.0001032258064516129, | |
| "loss": 1.9806, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.4337349397590362, | |
| "eval_loss": 0.2806684672832489, | |
| "eval_runtime": 46.0655, | |
| "eval_samples_per_second": 10.832, | |
| "eval_steps_per_second": 2.714, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.5943775100401605, | |
| "grad_norm": 2.434161424636841, | |
| "learning_rate": 9.247311827956989e-05, | |
| "loss": 1.8809, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.5943775100401605, | |
| "eval_loss": 0.27644357085227966, | |
| "eval_runtime": 44.1705, | |
| "eval_samples_per_second": 11.297, | |
| "eval_steps_per_second": 2.83, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.7550200803212852, | |
| "grad_norm": 3.0396993160247803, | |
| "learning_rate": 8.172043010752689e-05, | |
| "loss": 1.8046, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.7550200803212852, | |
| "eval_loss": 0.27500617504119873, | |
| "eval_runtime": 44.3244, | |
| "eval_samples_per_second": 11.258, | |
| "eval_steps_per_second": 2.82, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.9156626506024095, | |
| "grad_norm": 2.961737632751465, | |
| "learning_rate": 7.096774193548388e-05, | |
| "loss": 1.929, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.9156626506024095, | |
| "eval_loss": 0.2731240689754486, | |
| "eval_runtime": 43.8996, | |
| "eval_samples_per_second": 11.367, | |
| "eval_steps_per_second": 2.847, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0642570281124497, | |
| "grad_norm": 2.265627145767212, | |
| "learning_rate": 6.021505376344086e-05, | |
| "loss": 1.5674, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0642570281124497, | |
| "eval_loss": 0.2714921832084656, | |
| "eval_runtime": 45.3105, | |
| "eval_samples_per_second": 11.013, | |
| "eval_steps_per_second": 2.759, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.2248995983935744, | |
| "grad_norm": 2.743725061416626, | |
| "learning_rate": 4.9462365591397855e-05, | |
| "loss": 1.5314, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.2248995983935744, | |
| "eval_loss": 0.28298166394233704, | |
| "eval_runtime": 47.8967, | |
| "eval_samples_per_second": 10.418, | |
| "eval_steps_per_second": 2.61, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.3855421686746987, | |
| "grad_norm": 3.021275043487549, | |
| "learning_rate": 3.870967741935484e-05, | |
| "loss": 1.5405, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.3855421686746987, | |
| "eval_loss": 0.2852790653705597, | |
| "eval_runtime": 47.6731, | |
| "eval_samples_per_second": 10.467, | |
| "eval_steps_per_second": 2.622, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.5461847389558234, | |
| "grad_norm": 2.859989881515503, | |
| "learning_rate": 2.7956989247311828e-05, | |
| "loss": 1.4902, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.5461847389558234, | |
| "eval_loss": 0.28039097785949707, | |
| "eval_runtime": 49.8843, | |
| "eval_samples_per_second": 10.003, | |
| "eval_steps_per_second": 2.506, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.7068273092369477, | |
| "grad_norm": 2.8204994201660156, | |
| "learning_rate": 1.7204301075268818e-05, | |
| "loss": 1.4331, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.7068273092369477, | |
| "eval_loss": 0.27951765060424805, | |
| "eval_runtime": 49.9628, | |
| "eval_samples_per_second": 9.987, | |
| "eval_steps_per_second": 2.502, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.8674698795180724, | |
| "grad_norm": 2.8627824783325195, | |
| "learning_rate": 6.451612903225806e-06, | |
| "loss": 1.4442, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.8674698795180724, | |
| "eval_loss": 0.280029833316803, | |
| "eval_runtime": 49.9016, | |
| "eval_samples_per_second": 10.0, | |
| "eval_steps_per_second": 2.505, | |
| "step": 180 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 186, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 6.632546051948544e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |